chfritz

"Real API calls, real latency, real bills. Caveats are stated, not buried." -- Gosh, who talks like this? No human would. Claude does. Wannable confident, arrogant, and super annoying. Stated, not implied. ;-)

very cool! I'd be curious how some songs I know sound like, e.g., Stairway to heaven, nothing else matters, etc.

We've added support for VR controllers and mouse gestures to our Transitive Remote Teleop capability. This means you can now control your robotic arms over the internet from anywhere in the world (with live video and low latency).

New documentation re. arms: https://transitiverobotics.com/caps/transitive-robotics/remo...

Full description: https://transitiverobotics.com/caps/transitive-robotics/remo...

Looking forward to your feedback.

Congrats on the launch! At Transitive Robotics we make full-stack modules so robotics companies like you can build their operational systems faster and with greater quality. We have a number of modules you might be interested in, e.g., for remote video-streaming and tele-op/-assist: https://transitiverobotics.com/caps/. Let me know if you want to chat.

Very much besides the point. The point being that automation often enables other things, not just making the thing they automate directly more efficient.

"So if you can't collect enough data before building, and simulation won't give it to you, where does the data come from? Goldberg's answer is what he calls the data avalanche: build a working system using engineering plus some learning, ship it, then let production generate the training data." -- agreed! This is also the approach taken by Field AI if I'm not mistaken. Rather than calling it the data avalanche, you could also just describe it as "learning by doing".

nice idea and setup, but it doesn't work. I can request control, it will say "you are next" or similar, and then right away "time's up, return to games". I never get control.

Another one? What's wrong with the one by IEEE? https://robotsguide.com/robots

love the synthwave soundtrack! So full of action and out of style for chess, it's hilarious.

great idea! Now, you are running into the same issue Google Trends had to solve: term disambiguation. For instance, "atom" is ambiguous in a comparison of editors like this: https://hackernewstrends.com/?q=sublime&q=atom&q=vscode. Given LLMs it might be possible to use an embedding vector (with context) instead of a text string for indexing, and if you do, this problem might go away.

how is this different from using puppeteer to load the page and save the DOM as HTML?

A script or program is not a robot.

Almost definitely 100% teleoperated -- the robot might move between places autonomously, but all the manipulation tasks, i.e., the actual cleaning, is not something anyone yet knows how to do autonomously reliably. And that's OK. However I'd like to know whether the data you gather from these cleanings is shared with other companies (who would be willing to pay a lot of money for it in order to train their own physical AI models).

What a silly idea. Technology moves 20x faster than the legislature. There is no point in even trying this. By the time any suggestions would get implemented and (all?) signs replaced, they would already no longer be necessary -- like, 15 years past being necessary.

Makes a lot of sense, and seems better executed than AWS's pathetic RoboMaker attempt 8 years ago or whenever that was. Was looking for the team but there is no linkedin link on your page.

Obligatory: https://xkcd.com/927/

The problem is not that there is no standard -- there is, it's called ROS. The problem is that manufacturers are not interested in opening their robots up to that degree.

Also, REST is a terrible idea for interacting with robots. REST is call-response based. But with Robots a pub/sub (like ROS or MQTT) or a blackboard data architecture (like what we used in Transitive Robotics) is much more efficient and natural.

Foxglove is not the only name in town. There are many, Transitive Robotics, the company I'm building is one of them. Different from Foxglove we are much more focused on live-remote monitoring and control, e.g., we have a pretty popular remote teleoperation module: https://transitiverobotics.com/caps/transitive-robotics/remo... You can find all the other modules we're currently offering here: https://transitiverobotics.com/caps/ The platform itself is and remains open-source.

We are thrilled to announce a new major version of Transitive, the open-source framework for full-stack robotics. Version 2.0 adds significant new integrations and features: storage of historic and time-series data in ClickHouse, visualization in Grafana, and custom alerting via Alertmanager. Some of our capabilities, like the free Health Monitoring capability, already use these features, providing significant added value to robotics companies with growing fleets.

This is absolutely awesome. Thanks for sharing! I would love to chat more with you. For context: we make a remote teleoperation solution for robotics. It's mostly used for mobile robots, but we've been getting a lot of inquiries regarding teleoperation for manipulation, so I've been learning more about this, in particular regarding the question of speed. I really appreciate these results!

Impressive!

Agreed. Feels icky. Made me want to leave the page as quickly as possible again.

"the funding all these startups are getting should allow them to scale their methods 10x-100x.." .. "Therefore, we might soon see a ChatGPT moment in robotics" -- I don't think so and no, the second statement is NOT entailed by the first. Why would it? Because 100 is a big number? Do you have any idea how much more data LLM needed to be trained for a GPT3 level compared to the data available for robot training right now, and how low dimensional the space is in which LLMs operate compared to robots?

"My intuition is there's 40% chance we will see it this year" -- again, why? Don't you realize that people have been working in robotics for 65 years, and these people don't live under a rock either. They knew about GPT3 because 2023. So why is it NOW less then 10 month you think that this breakthrough will happen?

> 10 patents in 4 days

How can you seriously state that as a positive? Are you trying to exemplify the new AI-patent-generation-mill? No one in their right mind can expect a person to make 10 meaningful inventions in 4 days.

How many more times do you want to post this?

Working with nine different robotics companies over the course of 10 years has taught us a thing or two about designing robotic full-stack architectures. All this experience went into the design of Transitive, the open-source framework for full-stack robotics. In this new mini-series of blog posts we dive into the three core concepts of the framework.

We think that once you've experienced the magic of full-stack encapsulation, you'll not want to go back to the old ways of writing distributed code, esp. in robotics.

Sure, let's instead encourage them to reinvent the wheel (TF, simulation, architecture, navigation, localization, SLAM, 2D map representation, etc.) that seems much smarter. ROS might be like democracy ("the worst form of government except for all the others we've tried"), but it's still the only stake in the ground around which we've manage to corral a community.

For this question specifically: there are many ways to approach robotics, incl. from the hardware side. But approaching it from the software (sim) side is not invalid, so I don't think down-voting my comment was warranted.

Go through the ROS tutorials, using a simulated robot at first, then buy a tiny AMR kit and do the same thing IRL. Once you have those basics, you can ask yourself into which direction to go next, the obvious choices being 1) deeper into the research of the new-fashioned (bi-)manual manipulation (arms), or 2) more into business and actually build a real-world application for your mobile robot (which will involve a lot of tinkering with hardware). And +1 on what brudgers said. It's a hobby, so have fun.

Reverse engineering is illegal in many cases. Aren't you afraid you might be automating the process for your users to get into (legal) trouble? Will your tool warn the user if they are about to violate laws?

Doesn't that depend on what a failure looks like? Failure to achieve the task (with no other side effects) is very different from accidentally killing the patient. Maybe also send a link to that study/article you are referring to?