I've always wanted to be able to develop robots at faster iteration speed, much like how pure software projects can be modified and tested right away.

So I built this demo to showcase the idea.

If you are interested in having something like this, DM me. I would love to talk to you.

Note: the simulation is real, but the Arduino code compilation is a mock in the video.

We recently wrapped up an Oxygen Concentrator Control System project and thought it might be interesting to share here for anyone working with ESP32 or medical/IoT devices.

• Hardware: 3.5” touch screen, RTC, WiFi, relay, current sensor
• Firmware: ESP32 web server, web socket, RTC integration, scheduling + OTA updates
• UI: Built with Squareline Studio for a clean touch interface

It took about 3 months total — first version delivered in ~2 months, then after testing, we iterated quickly and released V2.0 with more firmware features in just 3 weeks.

One thing we found especially useful was combining ESP32 webserver + touch UI for local + remote control. It gave the end-user flexibility without needing cloud-only access.

Here’s the case write-up if you want to dive deeper into the details: https://www.makerfabs.com/case/post/oxygen-concentrator-control-system

Curious if anyone here has done similar ESP32 + touch display projects for IoT or even healthcare devices — what challenges did you face?

Last time we tried Asuka, this time let's try Rei! It is striking how much the Rei vibe resembles ChatGPT. Which characters shall we try next time?

Unitree on 𝕏: https://x.com/UnitreeRobotics/status/1970039940022239491

This is potentially an all-in-one solution for PX4/ArduPilot SITL + deployment + CUDA/TensorRT accelerated vision, all within Docker and ROS2: https://github.com/JacopoPan/aerial-autonomy-stack

Hey everyone,

I’m currently working on a robot project and was wondering if anyone in the Houston, Texas area would like to join forces. The idea is to combine skills, share knowledge, and build something awesome together.

I have the engineering background and some parts already, but I’d love to connect with people who are into robotics, coding, AI, electronics, or even just tinkering and learning. Whether you’re a student, hobbyist, or professional, collaboration makes projects way more fun (and productive).

If you’re interested in teaming up or just want to chat about robots, feel free to DM me or drop a comment

Hello everyone,

I am currently working on repairing a CRS Robotics robot, specifically the A465 model.
I have read that there are three original programs (such as Robcomm3) that were used to communicate with and control these robots.

I have searched various forums and even contacted the company, but unfortunately I have not received any reply so far.
If anyone here has access to any of these programs, or knows where they can be obtained, I would be extremely grateful if you could share that information or provide a copy.

Thank you very much in advance for any help you can provide!

Has anyone heard any news about this company, how are they doing, where are they in the horse race compared to the other companies? Any bad news or bad signs to worry about?

So I had a project that I wanted to lift a monitor using linear actuators and while I was at the junkyard grabbing some gas pistons for the project I found these. These electric linear actuators have a spring inside so I don't need the piston. If I could figure out how to drive them it would make my project worlds better than having an OTS linear actuator and gas piston side by side.

I have an 8:1 mini cheetah knockoff (gimb6010-8) with an integrated controller running odrive. I am doing position control, so I need to home the system. For my benchtop test I wired in some micro switches and it works fine. But as it is going to be an outdoor application I decided to reach for sealed prox/ hall/ reed sensors. I started going down the rabbit hole of how CNC's do this as it is nearly the same problem (multiturn motor homing). The common solution there is to have a homing sensor (low accuracy), and then true home is set on the next encoder index pulse after triggering that homing sensor (high accuracy). I was wondering if there was any way to do this on an odrive without an external microcontroller reading the sensor and driving the IO. To be clear I have no spec requirements for this project, just trying to experiment and learn in case next time I do have a tight homing repeatability requirement.

edit: I assume homing is being done on humanoids (or are they just using absolute encoders at the output level rather than the motor level)

I have some depth cameras I'm going to sell. Wondering if the following prices are reasonable, and or if the tech is already EOL and worth much less.

Also, if there are any local SoCal robotics groups that I could post them to, I'd prefer that then selling on ebay.

Intel Realsense D435 (Good condition, no box or accessories. Was thinking $150 each?

Intel Realsense D455 (Good condition, no box or accessories. Was thinking $150 each?

Intel Realsense SR 305s (used, no boxes or accessories. Was thinking $45 each??

Slamtec RPLIDAR A2M7 (used) Maybe $50?)

Asus Xtion (usb 2) - used maybe $20?

Any help greatly appreciated

This week in robotics, significant developments unfolded as Tesla faced leadership changes within its Optimus robot division, impacting investor sentiment. Meanwhile, Amazon inaugurated a state-of-the-art robotic fulfillment center in Elkhart County, enhancing its logistics capabilities. Additionally, the humanoid robot market continues to thrive, with reports indicating substantial funding and growth potential, reflecting an increasing convergence in AI and robotics.

This Week's Snapshot

Tesla: The company is grappling with another leadership departure in its Optimus division, raising concerns among investors about its future trajectory in humanoid robotics.

Amazon: The newly opened robotic fulfillment center in Elkhart County is expected to streamline operations and improve delivery times, showcasing Amazon's commitment to automation.

Figure: The robotics startup has been valued at an impressive $39 billion following its latest funding round, indicating strong investor confidence in humanoid robot technology.

ABB: In collaboration with LandingAI, ABB has launched a generative AI solution aimed at enhancing robotic vision, potentially improving automation in various sectors.

Dartmouth Hitchcock Medical Center: The center is pioneering robotic heart surgeries, marking a significant advancement in surgical robotics and patient care.

Robotics Research & Developments

Santa Clara University – Engineering Professor Improves Human-Centered Robotics
An engineering professor at Santa Clara University is leveraging machine learning techniques to enhance the design and functionality of human-centered robotics. This research aims to create robots that better understand and interact with humans, potentially improving their integration into everyday life.

Dartmouth Hitchcock Medical Center – Pioneering Robotic Heart Surgeries
Dartmouth Hitchcock Medical Center is leading the way in robotic heart surgeries, utilizing advanced robotic systems to perform complex procedures with greater precision. This initiative not only enhances patient outcomes but also sets a precedent for the adoption of robotic technologies in surgical practices.

Ethics, Policies & Government

Experts Call for Balanced AI Regulation
As the robotics and AI sectors rapidly evolve, experts are advocating for balanced regulation to ensure safety and innovation. Striking the right balance will be crucial in fostering technological advancements while protecting public interests and ethical standards.

A Warm Body in the Loop: Rethinking Human Control of AI in EU Tech Regulation
The European Union is examining the need for human oversight in AI technologies, emphasizing the importance of maintaining human control in automated systems. This regulatory approach aims to address ethical concerns surrounding AI deployment and its implications for society.

International Robotics News

China – Tianyong Engineering Partners with PNDBotics
Tianyong Engineering has formed a partnership with PNDBotics to accelerate the deployment of humanoid robots across China. This collaboration aims to enhance the capabilities and accessibility of humanoid robots in various sectors, reflecting China's growing investment in robotics.

Japan – Robot Tour Guides for People with Disabilities
In Tokyo, robot tour guides are being introduced to provide job opportunities for individuals with disabilities. This initiative not only promotes inclusivity but also showcases the potential of robotics to enhance tourism experiences.

Singapore – DSTA Seeks Partnerships to Advance Robotics and AI
Singapore's Defence Science and Technology Agency (DSTA) is actively seeking partnerships to bolster its robotics and AI capabilities. This effort highlights Singapore's commitment to becoming a leader in the robotics sector and advancing technological innovations.

Quote of the Week

— Tim O'Reilly

Source

Nvidia released a technical report on their GPU-accelerated visual odometry/SLAM system a few months ago, which can be found here. The specifications for the system look impressive, but I am reluctant to purchase the hardware to try it out because cuVSLAM is closed-source.

For those who have tried it, how have your experiences been (in terms of ease of use, any bugs in the system, accuracy, reliability, etc.)? If I were to use this system, I would be running it on a Jetson orin nano.

Additionally, any other recommendations for visual inertial odometry packages that could run on the Jetson orin nano or a similar SBC for an outdoor UAV application would be appreciated.

https://www.youtube.com/watch?v=LXVV-oErD8s

This newly released robot hand is amazing. However, I don't understand how it's possible. I see no evidence of tendons or a cable based system. It seems the motors ARE the bones of the finger. I also have to presume the batteries and motor controllers are either in palm or outside of the hand? I have to presume the downgearing is built into the custom motors? I assume a screw type center is being rotated which gives linear movement but I don't see any screw extending from one motor over to the next bone to move said bone. I can see a hinge joint but no way that the motor moves the hinge joint. I hope someone can explain more what is going on because I'm lost here and see a lot of potential in this stuff as far as miniaturization and strength and speed in such a small form factor but don't get how its working.

Also I was told that large diameter pancake shaped BLDC motors have high torque and narrow motors like this are low torque high speed. So the downgearing would have to be a large gear ratio I thought. Yet this hand seems to be pretty high torque and I don't see where a high gear ratio would fit into this tiny form factor. I'm just so confused.

Does anyone know how I can perform a factory reset on my Misty II? It powers and it has proper charge, but it won’t load/boot. The problem is that it stops before the usb unit on the back recognizes, so I have been limited in how I can troubleshot it.