[Shenzhen, September 12th] – Seeed Studio, a leading open AIoT hardware company, launches its new AI gadget SenseCAP Watcher on Kickstarter today. As the world’s first Physical LLM Agent for Smarter Spaces, SenseCAP Watcher is designed to bridge the gap between digital intelligence and physical applications.

SenseCAP Watcher marks a milestone in applying intellgence to interactions with spaces. Engineered with advanced hardware and software, the SenseCAP Watcher is not just a gadget but a comprehensive solution for enhancing spatial intelligence. Equipped with a camera, microphone, speaker, and touchscreen, SenseCAP Watcher understands and executes commands, comprehends the environment to trigger appropriate actions, offers human-like interactions, and integrates seamlessly with legacy systems while prioritising privacy.

Enhanced Monitoring for Spaces and Beyond

SenseCAP Watcher is adept at monitoring designated spaces, identifying specified targets, and spotting noteworthy events to deliver timely notifications. Simply speak to the Watcher, instruct it on what to monitor, and it will alert you upon occurrence of the specified events, which can be commanded via Push to Talk or through the SenseCraft App.

Advanced Scene Understanding

Leveraging on-device AI models, the SenseCAP Watcher can detect various objects and actions. By integrating with LLMs, it provides a deeper analysis of the scene, capturing detailed behaviors and statuses beyond basic target detection. This integration allows for precise event monitoring and reduced operational costs of using LLMs frequently.

Integration and Expansion

The SenseCAP Watcher can be integrated into various systems:

●It acts as a behavior sensor when connected to IoT platforms like Home Assistant, triggering tailored actions in diverse contexts.

●It can be added to Node-RED flows, allowing results to be streamed effortlessly across the internet.

●Compatibility with popular MCUs and SBCs like XIAO, Arduino, ESP32, and Raspberry Pi is ensured, facilitating the expansion into a broader range of intelligent applications.

Customization and Open Source Commitment

With the commitment of creating developer-friendly products, SenseCAP Watcher is fully open source and supports on-premise local LLM deployment. This ensures privacy and control over data and interactions. Users can customize the device further by adding new sensors via its I2C interface, which supports over 100 Grove sensors for multi-modal sensing capabilities.

The SenseCAP Watcher is now officially launched on Kickstarter from September 12 to October 12, 2024, with first 3-day exclusive Launch Special Offers starting from $79. Don’t miss this limited-time offer. Get your Watcher now!

About Seeed Studio

Founded in 2008, Seeed Studio has been a pioneering force in Open Hardware, enabling over half a million users to bring digital projects to life. Specialized in AI-enabled sensor networks and edge computing, Seeed Studio provides robust modules and devices, complemented by a comprehensive Fusion service for prototyping, production, and promotion. With a steadfast commitment to open source and community collaboration, Seeed Studio’s platforms like SenseCraft software suites facilitate inclusive and innovative digital solution development, fostering a democratized approach to technology.

The Month of XIAO: August 2024

Howdy,

I know this comes a bit late, but here is the debut episode of the “Everything About XIAO” Newsletter, wrapping up all the cool updates, new products, and cool projects, all happened in the amazing month of XIAO: August 2024.

Hope you’ll have a good read!

-Lily

01 New Products

2 New XIAO in 1 Month: Meet XIAO RP2350 and XIAO RA4M1

3 New Add-ons & Accessories for XIAO

How About a tinyML+LLM-Powered AI Device for Your XIAO?

Ever wanted to enhance your XIAO with the power of AI? Now you can with SenseCAP Watcher, the world’s first physical AI agent built on tinyML and LLM. SenseCAP Watcher helps you monitor anomalies and act accordingly based on voice commands. With its camera, mic, and environmental sensors, it can see, hear, and understand. By connecting SenseCAP Watcher to XIAO via UART, you can elevate your XIAO projects—whether in home automation, robotics, or other fields—to the next level. Discover how you can unlock the full potential of SenseCAP Watcher with XIAO! PS. Watcher will be live on Kickstarter on Sep 12, 2024. For all XIAO fans, we have a special package! Back Watcher and you can add an XIAO ESP32C6 for just $1.

02 Cool Community Projects

03 Creations Built on XIAO

04 XIAO in GitHub Repos

Microsoft DeviceScripte Supports XIAO ESP32C3

That’s all for this debut episode. Wanna have this monthly newsletter delivered to your email? Subscribe to our Everything About XIAO Newsletter NOW by clicking the image below.

It’s a monthly newsletter centering around the beloved Seeed Studio XIAO. We’ll cover the following info: 

 Cool Projects from the Community to get inspiration and tutorials
 Product Updates: firmware update, new product spoiler
 Wiki Updates: new wikis + wiki contribution
 News: events, contests, and other community stuff

Our renewed livestream “Making Next Gadgets” is back! Each livestream will feature a different tech theme, with our product dev teams sharing the latest product updates, project demos, and exclusive deals. Join us on a journey through the tech revolution happening right now!

Stay up-to-date with the latest product releases from Seeed. Discover new tools, platforms, and services designed to streamline your workflow and accelerate innovation in your projects.

Get the inside scoop on recent events and news. Gain valuable insights, key takeaways, and highlights from industry experts and thought leaders at the forefront of ML innovation.

Celebrate the creativity and ingenuity of our community members with featured projects and ideas in machine learning. From open-source libraries to DIY solutions, there’s no shortage of inspiration to be found among our talented community.

01. uPhone, An AI Wearable for Better Interaction, Powered by XIAO ESP32S3

Created by visionary technologist Cayden Pierce, uPhone is not just another AI wearable – it’s a paradigm shift in human-computer interaction. This groundbreaking device takes the form of a necklace, with a display that faces outward, visible to everyone engaged in conversation with the wearer. It’s a wearable computer that bridges the gap in human augmentation by making AI assistance, real-time translation, and information sharing a collective experience. Check it out.

We successfully hosted a workshop in Cameroon (Douala and Yaoundé)! The event was a collaboration between TmeEducation_Cameroon, Robotsave, Inchtech, Coveriot, LightGroup , KandaWeather and HardwareCommunity. It occurred at Light Group, Inchtech, and Coveriot centres in Douala and Yaoundé. The month was filled with innovation, learning, entrepreneurship, personal development, and collaboration.

A Great Attendance

Workshop Content

We are grateful to have had over 49 attendees from various milieus join us, including Students from different universities, makers and Startups, Telecom, Electrical and Electronics Engineers, a Young Researcher from the Ministry of Science and Development, Personnel from the Ministry of Post and Telecommunications, MTN Cameroon Telecom provider,  and the Telecom Regulator Agency (l’ART). 

This diverse group of attendees came together to explore the world of IoT and its different cases using IoT base LoRa tools provided by Seeed Studio such as SenseCAP LoRa Node industrial sensors and Gateways  with XIAO ESP32S3 Sense sense microcontrollers, which are powerful tools for rapid prototyping and real life project implementation. Additionally with the use of XIAO ESP32S3 Sense for a great possibility of AI vision model development. 

Impact and Future

Unlock the Future of IoT in Cameroon

The workshop was conducted in English and French for 6 days in Douala and Yaounde(morning and evening section). Participants learned and had hands-on experience with various IoT tools, including demonstrations and interactive projects such as a small aquaponics system and smart plant monitoring. This allowed attendees to directly experience the capabilities of this versatile industrial IoT LoRaWAN sensors and Gateways. Real-world use cases and best practices for implementing IoT-based LoRa projects were also presented.

Program details:

• Day 1: Introduction to the World of IoT and LoRaWAN with its Network Architecture, Sizing with the antenna characteristics to have a wide area coverage and a showcase of the projects performed recently in Cameroon: launching into space of Weather balloon radiosonde, installation of Lora gateways and SenseCAP Weather Station in Douala and Dibamba

• Day 2: Setting up, configuration and registration of  industrial sensors (SenseCAP Temperature-Humidity and moisture LoRa Node sensor), SenseCAP LoRa Gateway to SenseCraft and TTN,

• Day 3: Create a LoRa Local network between the LoRa Gateway and LoRa Node: XIAOESP32S3 microcontroller embedded with WIO-LoRa-E5 for Transmission, DHT11(Temperature, Humidity) and Soil capacity sensor and Tmedecation Arduino board which an educational board base on Arduino embedded with sensor such as light, Temperature and Potentiometer where we plug a Grove shield together with Lora-E5 transmission module.
• Day 4:  Installation and Introduction to  Node-Red that will help create a  Dashboard interface to visualize sensor data collected from the Diffrent loRa Node mentioned in Day 3.
• Day 5: Installation and Introduction to Mqtt mosquitto and Ngrock, to create a public broker that will enable the visualization of the Node-Red Dashboard over the Internet using Mqtt protocols, after performing participating dive into building projects: Aquaponics system, smart green plant pot.
• Day 6: Project Presentation and showcase at the National University of Post Telecommunications Information and Technology, 

This was a successful First Ranger IoT LoRaWAN workshop delivered in both French and English in Cameroon and this is a great testament to the growing interest in innovative technological solutions in the regions of Cameroon. At Seeed Studio, we are committed to continuing to support makers, Developers, and professionals on their path to innovation.

We thank everyone who participated and partners in the workshop event for its success. We look forward to organising more workshops and events, reaching more people and sharing the incredible joy of Technology with the work of Seeed Studion worldwide.

Stay tuned for our upcoming activities and keep innovating with us, Together we will go further!

Launching of the Monthly Campaign

One of the most exciting moments of the event was the announcement of the Monthly Campaign to Scale Up Your AI Gadget launched by Seeed Studio. This campaign is designed to help innovators transform their ideas into viable businesses. It offers comprehensive support from the conceptualization phase to commercialization, leveraging the expertise and resources of Seeed Studio.

We invite all attendees and the broader community to enable notifications for the program launch. We are excited to see how this program can drive new ideas and projects in the region.

As smart home automation continues to grow in popularity, choosing the right hardware to run Home Assistant is crucial. Many people are curious about Home Assistant Green or Raspberry Pi, here’s a quick recommendation based on your needs:

• Choose Home Assistant Green if you prioritize a quick, stable, and straightforward smart home setup with minimal maintenance.
• Choose Raspberry Pi if you prefer to customize your system, enjoy optimizing performance, and are looking for a cost-effective, versatile platform.

Now, let’s dive into the detailed comparison to help you make an informed decision.

Home Assistant Green

Key Features

1. Ready-to-Use Home Assistant Device: Just connect the power and network cables, and you’re ready to go. The mobile apps make the setup process smooth and easy, enhancing your smart home experience.
2. Data Localization: Keep your data private with exclusive control and local storage enabled by default.
3. Continuous Improvement Through Open Source: Enjoy monthly updates with new features and enhancements, thanks to the active open-source developer community.

Use Cases

• Smart Home Control
• Home Security
• Home Automation Tasks
• Energy Management

Pros

• Low entry barrier, ideal for beginners
• Compact design, easily integrates into home environments
• Affordable

Cons

• Expansion requires external devices (e.g., Zigbee or Z-Wave support, you can add support for Zigbee 3.0 and Matter by purchasing the Connect ZBT-1 USB Stick ).

Summary

Raspberry Pi

Raspberry Pi is a versatile microcomputer platform, beloved by tech enthusiasts for its flexibility and customization options. It can run Home Assistant and handle complex automation tasks.

Key Features

1. High Customizability: Choose your operating system, install various software packages, and use it for projects beyond smart home control.
2. Strong Community Support: Raspberry Pi has a large global user community that offers extensive tutorials and support resources, helping you tackle any challenges.
3. High Expandability: You can enhance storage performance with an SSD or extend system functionality with external modules.

Use Cases

• Custom Smart Home Systems
• Education and Learning Platforms
• Embedded Systems Development

Pros

• Affordable, great for budget-conscious users
• Highly customizable, perfect for tech enthusiasts
• Enhances reliability and performance by adding an SSD

Cons

• More complex setup, requires more time and technical skills
• SD cards are prone to failure; long-term use requires SSD for stability

Summary

Raspberry Pi is a highly flexible microcomputer platform suitable for those who enjoy DIY and technical customization. While the initial setup may require more time and technical knowledge, its affordability and expandability make it an ideal choice for smart home projects. By adding an SSD, you can further enhance its performance and reliability, ensuring the system performs well over the long term. You can now purchase the Raspberry Pi 5 8GB from Seeed for $80.

Comparison Summary

Conclusion: Which One Should You Buy?

Deciding between Home Assistant Green and Raspberry Pi depends on your needs, budget, and technical expertise. Here is a more detailed recommendation for each device:

Home Assistant Green

If you are a newcomer to smart home systems or want a quick setup and ready-to-use solution, Home Assistant Green is an excellent choice. Its user-friendly design allows you to start controlling your smart home devices without the need for extensive configuration. The device is constantly updated with new features and improvements by the open-source community, ensuring that your system stays current and secure.

For users who need to manage multiple brands and platforms in one app, Home Assistant software offers seamless integration across Apple HomeKit, Google Home, Amazon Alexa, and more. Although Home Assistant Green has a slightly higher initial cost, its built-in privacy features and ongoing community support make it a solid investment for those seeking a stable, low-maintenance smart home solution.

Raspberry Pi

On the other hand, Raspberry Pi is best suited for those who enjoy hands-on customization and have the technical skills to configure their system. If you appreciate the flexibility to tailor your setup and are willing to invest time in optimizing it, Raspberry Pi offers endless possibilities. Beyond just a smart home hub, it can serve as a powerful microcomputer for various applications.

Raspberry Pi‘s expandability is one of its greatest strengths. You can add an SSD for improved performance and durability, as well as connect additional modules to extend functionality. While the initial setup may be more complex, its affordability and flexibility make it an excellent choice for budget-conscious users who want to maximize their hardware investment.

Whichever platform you choose, you can build the ideal smart home system tailored to your needs and budget. We hope this comparison helps you make a more informed decision!

You May Also Like:

• • Home Assistant Green vs Yellow Which is Better
  • Top Raspberry Pi AI Projects for Beginners and Up

Imagine you’re in a place with no 4G, 5G, or WiFi. How would you stay safe and keep in touch with your friends? You might think of walkie-talkies or satellite phones, they can indeed be useful in certain situations. But today, I want to share with you our new T1000 meshtastic tracker that is cheaper, uses less power, covers more range, and is easy to use called SenseCAP Card Tracker T1000-E. With the T1000-E, you can easily build open-source, off-grid, decentralized and FREE LoRa mesh network for reliable long-range, low-power communication, real-time positioning, as well as getting some envrionmental data.

Meet SenseCAP Card Tracker T1000-E

SenseCAP Card Tracker T1000-E, the world’s first IP65-rated Meshtastic device with a nifty card size, effortlessly fits in your pocket or attaches to your assets, and works seamlessly with the Meshtastic LoRa mesh network. It embeds Semtech’s LR1110, Nordic’s nRF52840, and Mediatek’s AG3335 GPS module, providing a high-precision, low-power positioning and communication solution. With 2 onboard sensors, you can also get the temperature and lightness data as well.

Key Features of T1000-E

• IP65-Rated: The T1000-E Meshtastic tracker is tough enough to handle harsh environments no worries about damage, perfect for outdoor adventures. This makes it stand out from other Meshtastic devices.
• Real-Time Positioning: With the Mediatek AG3335 GPS chip, it offers high-precision positioning services and shares your location in real-time.
• No-Code Onboard: Thanks to the amazing Meshtastic, you can just pair it with a phone using the Meshtastic app via Bluetooth, and the T1000-E becomes your off-grid communication tool, no coding needed.
• Expandable Interfaces: Designed with four pogo pins, it supports USB interface for DFU (Device Firmware Upgrade), serial logging, and API interface, simplifying device management, debugging and development.
• Slim Card Size, Perfect for Carry-on Applications: Its dimensions are just 85mm x 55mm with a thickness of 6.5mm. It occupies minimal space and is easy to conceal, making it perfect for situations where reducing weight or maintaining discretion is important.

Since the T1000-E is based on Meshtastic, it comes with all the features Meshtastic has.

• Open-Source Free LoRa: It works with the Meshtastic open-source mesh networking protocol, ideal for free long-range and low-power communication.
• Direct P2P Communication: Any two T1000-Es can communicate directly over a distance of 5-10 km.
• Off-Grid & Decentralized: No signal or gateways needed. You can create a free mesh network with any T1000-E, even mix it with other Meshtastic devices. Any of these can act as a repeater to extend your message range.
• Encrypted Communication: Only the target node can view the messages, keeping your chats private.

What’s more, with its IP65-rated enclosure, it offers added durability for all these scenarios; with its built-in GPS, it provides real-time, high-precision positioning to ensure you’re always safe and easily located; With its low-power consumption, it offers up to 2 days of battery life, giving you plenty of time to enjoy your activities or handle an emergency situation.

What is Meshtastic?

For those unfamiliar with Meshtastic, it is an open-source mesh communication protocol based on LoRa, which enables the creation of a low-power, long-range, off-grid communication network.

You might have noticed that our T1000 series already includes LoRaWAN and Sidewalk trackers, so why are we releasing a Meshtastic version? Meshtastic, as an emerging network protocol that uses LoRa as its backbone and offers a wide range of possibilities. Its open-source nature, decentralization, encrypted communication, and cost-effectiveness make it incredibly versatile. That’s why we’ve introduced the T1000-E.

Differences between LoRaWAN®, Sidewalk, and Meshtastic

Meshtastic is better suited for flexible, independent, peer-to-peer communication scenarios, while LoRaWAN®is ideal for large-scale, centrally managed IoT applications. Amazon Sidewalk, is designed to extend the connectivity range of smart home devices using low-power, community-based networking. If you’re looking for a solution that allows you to stay connected in areas without network infrastructure or where network conditions are poor, the SenseCAP Card Tracker T1000-E is perfect for you.

Typical Meshtastic Applications

So, what are the typical applications of Meshtatsic? Here are some common usages shared by passionate community members on Reddit about “What do you use Meshtastic for?” .

1. calinet6: “Making new friends nearby who are equally nerdy by virtue of opting into this.”
2. Jcw122: “‘Talking’ with friends at big festivals”
3. spotcheck001: “We use it for family to communicate when we camp. Works out great.”
4. Stroniu44: “Combined with ATAK to localise and communicate during exercise with other tanks from my platoon, especially exchange kmz data”
5. jp_bennett: “Vehicle tracking across town, as well as emergency comms for emergency situations.”
6. AnderlAnduel: “I share weather station data between home automation sytems with my neighbors (mqtt2mqtt bridge). This way we are independent of Internet services.”

What You Can Do with SenseCAP T1000-E?

What they’re sharing are exactly the applications that we suggest you to start your Meshtastic journey with SenseCAP T1000-E:

• Outdoor Adventures
  • Communication Without Signal: Allows team members to stay in touch and share locations in areas without cellular service.
  • Team Coordination: Facilitates safe distance maintenance and collaboration among adventure groups.
• Remote Events and Festivals
  • Participant Communication: Ensures smooth communication at events where networks may be unavailable or overloaded.
  • Emergency Contact: Provides an alternative communication channel for alerts and assistance.
• Asset Tracking
  • Real-time Location Monitoring: Tracks goods in remote areas, ensuring continuous monitoring.
  • Valuable Asset Management: Use GPS to monitor changes in the item’s location and ensure its security.
• Search and Rescue
  • Coordinated Search Efforts: Enables search teams to share locations and optimize strategies.
  • Communication in Signal-deprived Areas: Supports communication among rescuers in challenging environments to enhance mission efficiency.

What the Meshtastic Community Says about SenseCAP T1000-E

Last but not Least, Spoilers!

SenseCAP T1000-E is the second addtion to our Meshtastic product lineup (the first one is the Wio Tracker 1110 Dev Kit for Meshtastic, offering more freedom for developers to hack and custom their Meshtasitc Tracker). But we won’t stop here. Please expect to see more Meshtatic powered products ranging from dev kits and industrial ready devices of pet tracker, dev board with XIAO, weather sensor and so on—coming up soon. If you have any other ideas, suggestions about our Meshtatsic-compatible products, feel free to share them with us in the comment!

One more thing! We more and more business scenarios are starting to use Meshtastic solutions. If you’re interested in collaboration or in a customized design, like white-label, function removal or add-ons, custom enclosure, or if you have even better customization ideas to turn the T1000-E into a product that perfectly meets your personal needs, please feel free to contact us at iot[at]seeed.cc.

In the world of smart homes, selecting the right hardware is crucial for ensuring stable and efficient system performance. Home Assistant, an open-source smart home platform, is beloved by developers and users worldwide, and its associated hardware continues to evolve and improve.

This article will delve into several hardware products suitable for Home Assistant, including Home Assistant Green, Yellow, Home Assistant SkyConnect / Connect ZBT-1 USB Stick, mmWave Human Detection Sensor Kit, Raspberry Pi 5 8GB, and other recommended products, to help you build your ideal smart home system.

Home Assistant Green and Home Assistant Yellow: Simple and Efficient Smart Home Control Centers

Home Assistant Green

• Entry-Level Model: Ideal for beginners and average users, Home Assistant Green provides a straightforward solution for most smart home needs. It offers essential features and functionality, making it easy to set up and use right out of the box.
• Capabilities: Despite its simplicity, it supports a range of smart devices and basic automation scenarios. It’s powerful enough for typical smart home setups but remains user-friendly, requiring minimal configuration.
• Ease of Use: The device is designed to be user-friendly, making it accessible for those who prefer not to spend much time on DIY hardware or complex setups. It’s perfect for users looking to get started with smart home automation without a steep learning curve.

Home Assistant Yellow

Home Assistant Yellow is the advanced version of the Home Assistant controllers, aimed at users with more demanding smart home needs. It features a more robust processor and additional expansion options.

• Advanced Model: Home Assistant Yellow is designed for those who need more power and flexibility. It includes enhanced hardware capabilities and supports expansion modules, such as the Zigbee module, to handle complex automation scenarios and manage a larger number of devices.
• Enhanced Features: This model supports advanced automation rules and can integrate with a wider variety of smart devices. Its powerful processor and additional memory make it suitable for handling more intricate setups and a higher volume of connected devices.
• Flexibility and Expansion: Home Assistant Yellow provides greater flexibility and expansion options compared to Home Assistant Green. It allows for more sophisticated configurations and automation, catering to users who require a more customized and comprehensive smart home system.

You can purchase the Home Assistant Yellow Kit with Power Supply version for $124 USD or the Home Assistant Yellow Kit with Power-over-Ethernet version for $135 USD today at Seeedstudio.

Read More: Home Assistant Green vs Yellow vs Blue

Home Assistant SkyConnect / Connect ZBT-1 USB Stick: Optimal Solutions for Zigbee and Z-Wave

The SkyConnect USB Stick is the ideal solution for expanding your Home Assistant system with seamless Zigbee connectivity and future support for Matter through firmware updates. By plugging this USB device into your host, you can easily integrate and control Zigbee smart home devices, with plans to support Matter in the future. Available now at Seeed for just $29.99.

• High Performance, Low Power: Powered by the EFR32MG21 Wireless SoC, the SkyConnect supports full-speed USB 2.0 with the included extender, offering high performance while maintaining low power consumption.
• Reliable Communication: With a transmission power of +20 dBm, it ensures robust and consistent connectivity for your smart devices, enhancing the reliability of your smart home network.
• Firmware Updates Made Easy: Over-The-Wire (OTW) support allows for easy and consistent updates, keeping your device up-to-date with the latest features and improvements.
• Dual-Protocol Operation: Simultaneously runs Zigbee and Matter protocols, providing versatility and compatibility with a wide range of smart home devices.

mmWave Human Detection Sensor Kit: Accurate Smart Home Sensor

The mmWave Human Detection Sensor Kit utilizes microwave technology to detect human presence through minimal movement. This advanced sensor excels in environments where traditional infrared sensors might fail, such as through walls or glass. Purchase it today for just $22.94 at Seeed.

• Precision Detection: It can accurately detect human movement, making it ideal for applications like security monitoring, energy-efficient lighting, and automated climate control. For instance, it can automatically activate lights or adjust the thermostat when it detects occupancy, leading to increased energy efficiency and enhanced home automation.
• Versatile Use: Its ability to function through obstacles means it can be used in various settings, including complex or hidden locations, providing reliable detection regardless of the physical barriers in the environment.

Raspberry Pi 5 8GB: The Ultimate Choice for DIY Users

Are you finding an home assistant server alternative to Home Assistant Green and Yellow?

The Raspberry Pi 5 8GB stands out as a top choice for those who enjoy DIY projects and want to build a customized Home Assistant system. Its powerful hardware and extensive connectivity options make it suitable for a wide range of smart home applications. You can buy it for only $80 at Seeed right now

• High Performance: With its advanced processing power and 8GB of RAM, the Raspberry Pi 5 can handle multiple tasks and complex automations without performance issues. This makes it ideal for users who want to integrate numerous sensors, cameras, and other devices into their smart home setup.
• Customization and Flexibility: It offers unparalleled flexibility, allowing you to tailor your Home Assistant system to meet specific needs, whether you’re implementing intricate automation rules or connecting a variety of smart devices. The Raspberry Pi 5 also supports a wide range of add-ons and peripherals, enhancing its functionality and adaptability for different smart home scenarios.

Other Recommended Home Assistant Devices

In addition to the above, the following devices are also well-suited for Home Assistant systems, further enhancing your smart home experience:

reTerminal Series

Touchscreen terminal devices based on Raspberry Pi, allowing Home Assistant to be installed on a palm-sized device or mounted on a wall for convenient control.

Odyssey

A mini PC with powerful processing and extensive interfaces, ideal for running high-performance applications like Home Assistant.

LinkStar

A mini router supporting Docker, where you can run Home Assistant Docker containers, offering strong network features.

XIAO ESP32C3

A compact microcontroller for creating custom smart home nodes, easily integrated with Home Assistant.

SenseCAP Indicator

A sensor hub displaying and managing environmental data, perfect for monitoring and automating responses.

Summary: Choose the Right Hardware Based on Your Needs

Selecting the right hardware is key to building an effective smart home.

• Home Assistant Green and Yellow offer plug-and-play solutions for various user needs.
• SkyConnect / Connect ZBT-1 USB Stick supports Zigbee and Z-Wave networks.
• mmWave Sensor Kit provides enhanced security and energy efficiency;
• and Raspberry Pi 5 offers limitless possibilities for DIYers.
• Additionally, reTerminal Series, Odyssey, LinkStar, XIAO ESP32C3, SenseCAP Indicator provide diverse hardware options to enhance your smart home system.

Whether you are a beginner or an experienced user, these devices can help you create a high-performance, stable smart home system.

The “Bathing Water Quality” project at Boscombe Seafront, Bournemouth, marks a significant advancement in coastal environmental monitoring, driven by IoTech’s deployment of the UK’s first 9-in-1 smart weather station. This initiative, part of BCP Council’s “Smart Place” program, leverages SenseCAP compact weather sensor and the SenseCAP S21000 LoRaWAN Data Transmission Unit to provide real-time, accurate environmental data crucial for maintaining high water quality standards. In collaboration with UnifAI Technology, Wessex Water, and Iknaia, this project not only enhances public safety but also sets a new standard for smart coastal management, serving as a model for other coastal regions across the UK. IoTech, as a solutions provider, played a pivotal role in ensuring the successful integration and deployment of SenseCAP devices

• Solution: Installation of a 9-in-1 smart weather station to provide real-time, accurate environmental data for enhanced water quality monitoring.

• Participating Partners: IoTech, BCP Council’s “Smart Place” initiative, UnifAI Technology, Wessex Water, Iknaia

• Seeed Products used: SenseCAP 9-in-1 Compact Weather Sensor, SenseCAP S2100 LoRaWAN DTU

• Industry: Environmental Monitoring, Coastal Management

• Solution Deployed in: Boscombe Seafront, Bournemouth, UK

Bournemouth, a renowned coastal town in Dorset, UK, is celebrated for its stunning beaches and vibrant tourism industry. In recent years, the town has focused on enhancing its environmental monitoring capabilities to ensure the safety and enjoyment of both residents and visitors. With an increasing emphasis on sustainable tourism and public health, BCP (Bournemouth, Christchurch, and Poole) Council launched the “Bathing Water Quality” project as part of its broader Smart Place initiative, which leverages cutting-edge technology to monitor and maintain high environmental standards.

Challenges

Bournemouth’s tourism industry relies heavily on the pristine condition of its beaches and coastal waters. However, maintaining water quality in such a popular destination poses significant challenges, particularly due to pollutants and the impact of weather conditions on water safety. Traditional methods of water quality monitoring were insufficient to provide real-time data, which is crucial for timely interventions and informing the public.

The challenge was to implement an IoT solution that could provide accurate, real-time environmental data to monitor water quality, predict risks, and ensure the health and safety of beachgoers. Additionally, the solution needed to integrate seamlessly with the existing digital LoRaWAN network infrastructure and provide accessible information to the public.

Solution

Why SenseCAP

Results of the Project

Partners

More Case Study

FAB24 Mexico, hosted by The Fab Foundation, Fab City Foundation, The Center for Bits and Atoms at MIT,Institute of Design and Technological Innovation (IDIT), and The Center for Innovation, Entrepreneurship and Business (CIEN) of Puebla, is a celebration of innovation and creativity bringing together makers from around the world. Seeed Studio was thrilled to be a part of this dynamic event as a key sponsor again (read more about Seeed at FAB23 Bhutan).

Here’s an in-depth look at our participation and exploration this year.

1. CEO Keynote: An Invitation to Co-invent the Needed Wheels

From August 5 to 9, the Fab Lab Conference & Symposium was held at IDIT, which was packed with workshops, talks, and panels exploring the role of digital fabrication in promoting social change. Our CEO, Eric Pan, delivered an inspiring keynote at the Technology 5.0 symposium, introducing AI integration and the Shenzhen ecosystem, highlighting the potential of AI to enhance IoT applications and how emerging technologies can be leveraged to create a sustainable future.

Other speakers include MIT professor Neil Gershenfeld, Daniele Ingrassia of Inmachines, Dale Dougherty of Make Community LLC, Antonio Quirarte of Hacedores and Alejandro Mayoral of Indigenous Friends Association (IFA), etc. Watch the recorded livestream of all symposium days here.

2.Interactive Workshops

We also organized several interactive workshops, which were a major highlight offering hands-on learning experiences that empowered participants to apply cutting-edge technology to real-world problems:

● How to Add AI to Almost Everything

Designed for AI beginners and Arduino enthusiasts, this workshop provided a comprehensive introduction to TinyML using our XIAO ESP32S3 Sense. Participants discussed the future of AI and the role of Large Language Models in enhancing device intelligence, experienced interactive demonstrations of TinyML applications in real-time, while engaging in sound and vision model training. At the end, participants received the Grove Vision AI V2 kits to continue their AI journey, making this workshop an excellent entry point for those looking to integrate AI into their projects.

Missed the workshop? Don’t worry! Click the button below to access the course materials.

● Co-create Solution for Hazard Response

This workshop, aligned with the Fab City Challenge’s mission, was a collaborative effort to develop innovative, sustainable solutions for hazard response. Participants explored various hazard scenarios and identified gaps in current security systems, learnt to used AIoT hardware and Node-RED to develop and implement reproducible hazard response solutions. They engaged in hands-on projects, such as building resilient local warning systems and deploying advanced hazard sensors.

Initially, we planned to host two workshops with the above two topics, but due to the overwhelming response, we had to schedule encore sessions and switch to larger venues to accommodate everyone. We are incredibly grateful for the enthusiasm and participation, which exceeded all expectations!

We will update the course slides soon. Please stay tuned.

Apart from our own workshops, we also sponsored the XIAO RP2040 and XIAO ESP32 Sense to several other workshops hosted by our community memebers.

3. Engaging in Fab City Challenge

Our engagement extended beyond the conference venue, as our colleague Crail Lyu participated in the Fab City Challenge: Greening Communities. This initiative, hosted by FabLab Ciudad Retoño, focuses on bee conservation to ensure biodiversity and food security. Together with the team members, they’ve developed :

• A data-based decision-making system utilizing SenseCAP series that monitors temperature, light, moisture, weight and bee counting to enhance both gardening and beekeeping.
• A productive pollinator garden to support the preservation and expansion of biodiversive oases within the city

They showcased the results of their effort through interactive presentations on August 3-4 at the Fab Festival. Learn more about their work.

4. #SeeedinMexico

● Meet Up: LET'S PARTY! AI Hardware & Chelas

On August 2, in collaboration with Hacedores Makerspace, we hosted a special meet-up filled with innovation, fun, and community spirit. Attendees enjoyed drinks, snacks, and exclusive Seeed Studio products, with our CEO Eric Pan shared the latest news from Seeed.

● Interviews by MoonMakers and Nada que Hacer

At FAB24, we had the absolute pleasure of meeting the amazing MoonMakers and Nada que hacer, the rockstar content creators in the gadget tinkering world. Our CEO Eric Pan has engaged in conversations with them, delving into innovative approaches to digital fabrication and the diverse applications across communities. Watch the video by clicking:

● Setting up the Co-create Solutuion for Hazard Response Infographic at IDIT

FAB24 Mexico provided an excellent platform for networking with fellow innovators, industry leaders, and enthusiasts. Our team engaged in meaningful discussions, exchanged ideas, and explored potential collaborations. Seeed Studio is proud to have been part of this extraordinary event. We are grateful to all the organizers, participants, and partners who made this event a success!

Stay connected with Seeed Studio for more updates and future events.

End Note

Hey community, we’re curating a monthly newsletter centering around the beloved Seeed Studio XIAO. If you want to stay up-to-date with:

 Cool Projects from the Community to get inspiration and tutorials
 Product Updates: firmware update, new product spoiler
 Wiki Updates: new wikis + wiki contribution
 News: events, contests, and other community stuff

Please click the image below to subscribe now!

Introduction

Back in 2019, the concept of the “MING” stack—comprising MQTT, InfluxDB, Node-RED, and Grafana—was first introduced on the Balena forums. The vision behind MING was simple: empower embedded developers by providing a robust foundation of pre-built technologies, eliminating the need to reinvent the wheel. This approach not only accelerates project development but also serves as an ideal entry point for those venturing into IoT application development.

Much like how the LAMP stack (Linux, Apache, MySQL, PHP) revolutionized web development by addressing the challenges developers faced, the MING stack is poised to do the same for IoT and embedded systems developers. With the continued evolution of IoT, the open-source MING stack emerges as a beacon, enabling the next generation of IoT developers to achieve unprecedented levels of productivity.

The MING Stack Components

Mosquitto

At the heart of the MING stack is Mosquitto, a lightweight MQTT messaging broker that excels at handling data exchange in IoT environments. MQTT’s highly scalable publish-and-subscribe protocol is capable of managing millions of messages per second, making Mosquitto an ideal choice for large-scale IoT applications.

InfluxDB

The backbone of IoT data management, InfluxDB is an open-source time series database tailored for trend analysis across various IoT devices. Whether it’s predictive maintenance for factory equipment or monitoring environmental conditions, InfluxDB’s ability to record and analyze time-based data solves a myriad of IoT data challenges.

Node-RED

Node-RED is a powerful, open-source tool that simplifies the creation of data flows for IoT devices. By enabling seamless connectivity between devices, data sources, and services, Node-RED empowers developers to build custom IoT applications that process and analyze data in real-time, all through an intuitive drag-and-drop interface.

Grafana

When it comes to visualizing IoT data, Grafana stands out as a leading open-source platform. Real-time monitoring is often the first application deployed in IoT systems, and Grafana’s graphing and dashboard capabilities make it an essential component of any IoT architecture.

Why MING Stack?

The MING stack’s versatility allows it to be deployed both on the edge and in the cloud, offering unmatched flexibility for IoT projects. With large, active developer communities and a relatively easy learning curve, MING enables rapid development and deployment of IoT applications, making it a go-to choice for developers aiming to stay ahead of the curve.

MING Stack on Seeed Devices

reComputer R1000

The reComputer R1000 is a robust, ready-to-deploy edge IoT controller built on the Raspberry Pi CM4 platform. Equipped with a quad-core processor, it supports a wide range of configurations, including 3 isolated RS485 ports, dual Ethernet, and optional wireless connectivity (4G, LoRa®, Wi-Fi/BLE, Zigbee). Ideal for smart building applications, agriculture, and smart automation scenarios, the reComputer R1000 can seamlessly integrate with cloud services like AWS using MQTT, while its support for multiple wired and wireless protocols ensures comprehensive data collection and management.

The reTerminal DM is an all-in-one Panel PC, HMI, and IIoT gateway powered by the Raspberry Pi CM4. Featuring a 10.1” IP65 front panel, it supports the Raspberry Pi-based software ecosystem and is equipped with rich interfaces, including CAN bus, RS485, RS232, and Gigabit Ethernet. Its powerful wireless communication capabilities (4G, LoRa®, Wi-Fi, BLE) enable seamless data transmission to the cloud. Additionally, its 10.1” IP65 front panel allows for on-premises data visualization, making it an ideal choice for field data collection and real-time monitoring.

reTerminal DM

The MING stack, combined with Seeed’s powerful edge controllers, represents a formidable solution for modern IoT applications. Whether deployed on the edge or in the cloud, MING enables developers to rapidly create, deploy, and scale IoT solutions, driving innovation and efficiency in an increasingly connected world.

Demo on MING Stack

Since the release of the Arduino Uno R4 in March 2023, we’ve received many requests to incorporate the powerful Renesas RA4M1 microcontroller (MCU) into the popular Seeed Studio XIAO form factor. We’re excited to announce that it’s finally here with the new Seeed Studio XIAO RA4M1!

Seeed Studio XIAO RA4M1 is a groundbreaking addition to our XIAO lineup, expanding the possibilities with the powerful chipset Renesas RA4M1. The XIAO RA4M1 maintains the beloved features of the Seeed Studio family: a thumb-sized form factor, all components surface-mounted with stamp holes on both sides, compatibility with Seeed Studio XIAO accessories, and open-source availability. It also boasts a 32-bit ARM® Cortex®-M4 running at up to 48 MHz with FPU, along with ample memory and flash storage.

What’s New in the XIAO RA4M1?

As the latest member of the Seeed Studio XIAO family, the XIAO RA4M1 features the same processor as the iconic Arduino Uno R4.

The Renesas RA4M1 group of MCUs uses the high-performance Arm® Cortex®-M4 core and offers a segment LCD controller and a capacitive touch sensing unit input for intensive HMI designs. The RA4M1 MCU is built on a highly efficient low-power process and is supported by an open and flexible ecosystem concept—the Flexible Software Package (FSP), built on FreeRTOS—and is expandable to use other RTOSes and middleware. The RA4M1 is suitable for applications where a large amount of capacitive touch channels and a segment LCD controller are required.

"Seeed Studio has been pioneering open-source hardware solutions that empower developers worldwide. We're excited for this latest collaboration, embedding our powerful RA4M1 Arm Cortex®-M4 MCU into their signature XIAO Series Dev Boards. We look forward to seeing what new and exciting edge and TinyML applications the developer community creates with this tiny, thumb-sized board," says Brad Rex, Senior Director of System Solutions at Renesas.

XIAO RA4M1 Key Features

●Popular Microcontroller Onboard: Powered by the Renesas RA4M1, a 32-bit ARM® Cortex®-M4 MCU operating at up to 48 MHz with FPU, with 256 KB of Flash memory and 32 KB of SRAM.

●Highlighted Onboard Resources: Includes a 14-bit ADC, 12-bit DAC, CAN BUS, USB 2.0, and an onboard RGB LED.

●Expanded IOs: Adds 8 new IO pins on the back, totaling 19 GPIOs, enabling more complex applications.

●Efficient Power Design: Offers 4 operating modes with power consumption as low as 45μA in deep sleep, and supports lithium battery charge management.

●Software Compatibility: Fully compatible with Arduino IDE for seamless project development and prototyping.

●Hardware Compatibility: Thanks to the extensive XIAO ecosystem, the XIAO RA4M1 supports a wide range of add-ons, including displays, LED matrices, Grove modules, CAN Bus, Vision AI sensors, and mmWave sensors.

●Powerful Security Features: Built-in hardware encryption, secure boot, key storage, and more to ensure application security.

Join Our Early Adopter Program

The XIAO RA4M1 is now available for pre-order at $4.99, with shipping expected in late August 2024. 

To assist developers in adapting their projects to the XIAO RA4M1 quickly, we’re launching an early adopter program. If you’re an author or maintainer of a popular open-source Arduino project on GitHub, or an influencer on platforms like YouTube or Tiktok or Instagram, make sure your users can run your code on the XIAO RA4M1. Join to receive 2 free boards and technical support to make your code compatible with this exciting new board.

Notes at the end.

Hey community, we’re curating a monthly newsletter centering around the beloved Seeed Studio XIAO. If you want to stay up-to-date with:

 Cool Projects from the Community to get inspiration and tutorials
 Product Updates: firmware update, new product spoiler
 Wiki Updates: new wikis + wiki contribution
 News: events, contests, and other community stuff

Please click the image below to subscribe now!

Image source: AI-generated

In modern computing, the CPU (Central Processing Unit), GPU (Graphics Processing Unit), TPU (Tensor Processing Unit), and NPU (Neural Processing Unit) each play crucial roles. Understanding the differences and applications of these processing units can help you make the best choice for your specific needs.

Overview of Different Processing Units

1. CPU (Central Processing Unit)

Image source: intel

Features and Uses

• Versatility: The CPU is the heart of a computer, capable of handling various tasks, from simple text editing to complex mathematical calculations.
• Serial Processing: CPUs excel at serial tasks, efficiently executing one instruction after another with each core.
• Multi-Core Support: Modern CPUs typically feature multiple cores, enabling them to handle multiple tasks simultaneously, thereby improving overall performance.

Suitable Scenarios

• Everyday computing tasks (e.g., document processing, web browsing)
• Software development and compilation
• Lightweight graphics processing and gaming

2. GPU (Graphics Processing Unit)

Image source: NVIDIA

Features and Uses

• Parallel Processing: GPUs have a large number of smaller cores that can handle thousands of parallel tasks, making them ideal for graphics rendering and matrix computations.
• High Bandwidth: GPUs possess high memory bandwidth, allowing rapid data transfer and boosting computational efficiency.
• Graphics Rendering: Originally designed for graphics rendering, GPUs are now widely used in scientific computing and machine learning.

Suitable Scenarios

• Graphics-intensive applications (e.g., 3D rendering, video editing)
• Deep learning and neural network training
• Data analysis and scientific computations

3. TPU (Tensor Processing Unit)

Image source: Seeedstudio

Features and Uses

• Specialized Efficiency: TPUs are hardware specifically designed by Google to accelerate machine learning tasks, particularly tensor operations.
• High Efficiency: TPUs offer higher efficiency and lower power consumption compared to CPUs and GPUs when handling specific machine learning tasks.
• Integrated Ecosystem: Seamlessly integrated with machine learning frameworks like TensorFlow, TPUs optimize the training and inference speed of deep learning models.

Suitable Scenarios

• Deep learning model training and inference
• Efficient AI application deployment
• Cloud-based AI computing (e.g., Google Cloud TPU)

4. NPU (Neural Processing Unit)

Image source: Seeedstudio

Features and Uses

• Dedicated AI Acceleration: NPUs are designed specifically for neural network processing, optimizing the execution of deep learning models.
• Low Power Consumption: NPUs have lower power consumption compared to CPUs and GPUs when executing AI tasks, making them ideal for edge computing and mobile devices.
• Fast Inference: NPUs can quickly execute neural network inference tasks, enhancing the intelligence capabilities of devices.

Suitable Scenarios

• AI applications on mobile devices (e.g., image recognition, voice processing)
• Edge computing devices (e.g., smart cameras, IoT devices)
• Efficient AI inference tasks

Differences Between TPU and NPU

Who are the Manufacturers of CPUs, GPUs, TPUs, and NPUs？

CPU Manufacturers: Intel, AMD, IBM, Apple, Qualcomm, ARM, etc.

GPU Manufacturers: NVIDIA, AMD, Intel, ARM, Imagination Technologies, etc.

TPU Manufacturer: Google, Coral (owned by Google), HAILO, etc.

NPU Manufacturer:HAILO, Samsung, Qualcomm, Apple, Huawei, etc.

How to Choose the Right Processor for Your Needs?

Choosing the appropriate processing unit—whether CPU, GPU, TPU, or NPU—depends on the specific requirements of your tasks and applications. Here’s a detailed guide to help you make an informed decision based on your needs:

Everyday Computing and Multitasking

• Recommendation: CPU
• Why: Versatility and efficiency in general-purpose computing.
• Use Cases: Web browsing, office applications, media playback, light gaming

Graphics Processing and Deep Learning Training

• Recommendation: GPU
• Why: Excels at parallel processing.
• Use Cases: 3D rendering, video editing, deep learning training, high-end gaming

Focused Machine Learning Tasks

• Recommendation: TPU
• Why: Designed for machine learning workloads.
• Use Cases: Training large-scale models, real-time inference, TensorFlow applications

Mobile and Edge AI Applications

• Recommendation: NPU
• Why: Low power consumption and efficient AI processing.
• Use Cases: Image recognition, voice processing, edge computing devices

Learn more: Raspberry Pi AI Kit(Hailo-8L) vs Google Coral USB Accelerator

Combining Different Units

• CPU + GPU: Use the CPU for general-purpose tasks and the GPU for parallel processing tasks like deep learning training and video rendering.
• CPU + TPU/NPU: Use the CPU for general operations and the TPU or NPU for specialized machine learning tasks, optimizing both efficiency and performance.

Choosing the right processing unit is crucial for optimizing the performance and efficiency of your applications. By understanding the specific strengths and applications of CPUs, GPUs, TPUs, and NPUs, you can make an informed decision that best suits your needs. 

Whether you’re engaged in everyday computing, deep learning, or edge AI applications, the right hardware can significantly enhance your productivity and results.

For a deeper dive into how these processing units compare and to visualize their use cases, check out this video

Our renewed livestream “Making Next Gadgets” is back! Each livestream will feature a different tech theme, with our product dev teams sharing the latest product updates, project demos, and exclusive deals. Join us on a journey through the tech revolution happening right now!

Stay up-to-date with the latest product releases from Seeed. Discover new tools, platforms, and services designed to streamline your workflow and accelerate innovation in your projects.

Get the inside scoop on recent events and news. Gain valuable insights, key takeaways, and highlights from industry experts and thought leaders at the forefront of innovation.

Celebrate the creativity and ingenuity of our community members with featured projects and ideas in machine learning. From open-source libraries to DIY solutions, there’s no shortage of inspiration to be found among our talented community.

01. Transferscope: Using Gen AI and XIAO RP2040 to Transform Real-World Images

Transferscope, developed by Christopher Pietsch, is a XIAO RP2040 and Raspberry Pi powered tool that allows users to directly use their physical environments, bypassing the need for predefined prompts or extensive digital input and forming a prompt that captures textures, patterns, and concepts from real-world objects and applying them to other surfaces. Check it out. 

It’s yet another exciting release from Raspberry Pi today (for the Raspberry Pi Pico 2 and RP2350 chipset)! And here at Seeed we’re excited to announce Seeed Studio’s latest release – the XIAO RP2350! This tiny board packs the capabilities of the Raspberry Pi RP2350 chip into the beloved XIAO form factor, making it one of the smallest dev board for MicroPython. XIAO RP2350 is already featured in the Powered by Raspberry Pi Product Catalogue among with other solutions based on RP2350.

As one of the smallest MicroPython-ready dev boards available, the XIAO RP2350 is perfect for space-constrained projects that require high performance. It has all the features of the RP2350, including dual-core, dual-architecture capability, allowing developers to switch between the dual industry-standard Arm Cortex-M33 cores @ 150MHz with FPU, and dual Hazard3 RISC‑V cores, security features like secure boot and an encrypted bootloader. Despite its small 21×17.8mm size, it still manages to include 19 GPIOs, an RGB LED, battery management, and more.

Best of all, the XIAO RP2350 is compatible with tons of add-ons through the XIAO ecosystem. You can easily add displays, sensors, motors, LED matrix, expansion boards and more. It supports MicroPython, C and C++ right out of the box too.

Some highlighted features:

• 19 GPIOs (8 more than previous XIAOs!) allowing more complex projects
• Powerful dual-core and dual-architecture (Arm and RISC-V) RP2350 chip with FPU for demanding applications
• Security features like secure boot and encrypted bootloader protect your code
• Runs MicroPython, C and C++ for maximum flexibility
• Extensive ecosystem of add-ons like Grove sensors, actuators, displays, LED matrix and more

The XIAO RP2350 is in stock now with a limited first batch of 400 boards on both our Webstore and AliExpress Store. Don’t miss your chance to be one of the first to try out this tiny powerhouse – check out our product page today. Let me know if you have any other questions!

Join Our Early Adopter Program

To assist developers in adapting their projects to the XIAO RP2350 quickly, we’re launching an early adopter program. If you’re an author or maintainer of a popular open-source Raspberry Pi project on GitHub, or an influencer on platforms like YouTube or Tiktok or Instagram, make sure your users can run your code on the XIAO RP2350. Join to receive 2 free boards and technical support to make your code compatible with this exciting new board.

Notes at the end.

Hey community, we’re curating a monthly newsletter centering around the beloved Seeed Studio XIAO. If you want to stay up-to-date with:

 Cool Projects from the Community to get inspiration and tutorials
 Product Updates: firmware update, new product spoiler
 Wiki Updates: new wikis + wiki contribution
 News: events, contests, and other community stuff

Please click the image below to subscribe now!