Tag Archives: box
Video Friday is your weekly selection of awesome robotics videos, collected by your Automaton bloggers. We’ll also be posting a weekly calendar of upcoming robotics events for the next few months; here’s what we have so far (send us your events!):
ARSO 2019 – October 31-1, 2019 – Beijing, China
ROSCon 2019 – October 31-1, 2019 – Macau
IROS 2019 – November 4-8, 2019 – Macau
Let us know if you have suggestions for next week, and enjoy today’s videos.
We are very sad to say that MIT professor emeritus Woodie Flowers has passed away. Flowers will be remembered for (among many other things, like co-founding FIRST) the MIT 2.007 course that he began teaching in the mid-1970s, famous for its student competitions.
These competitions got a bunch of well-deserved publicity over the years; here’s one from 1985:
And the 2.007 competitions are still going strong—this year’s theme was Moonshot, and you can watch a replay of the event here.
[ MIT ]
Looks like Aibo is getting wireless integration with Hitachi appliances, which turns out to be pretty cute:
What is this magical box where you push a button and 60 seconds later fluffy pancakes come out?!
[ Aibo ]
LiftTiles are a “modular and reconfigurable room-scale shape display” that can turn your floor and walls into on-demand structures.
[ LiftTiles ]
Ben Katz, a grad student in MIT’s Biomimetics Robotics Lab, has been working on these beautiful desktop-sized Furuta pendulums:
That’s a crowdfunding project I’d pay way too much for.
[ Ben Katz ]
A clever bit of cable manipulation from MIT, using GelSight tactile sensors.
[ Paper ]
A useful display of industrial autonomy on ANYmal from the Oxford Robotics Group.
This video is of a demonstration for the ORCA Robotics Hub showing the ANYbotics ANYmal robot carrying out industrial inspection using autonomy software from Oxford Robotics Institute.
[ ORCA Hub ] via [ DRS ]
Meet Katie Hamilton, a software engineer at NASA’s Ames Research Center, who got into robotics because she wanted to help people with daily life. Katie writes code for robots, like Astrobee, who are assisting astronauts with routine tasks on the International Space Station.
[ NASA Astrobee ]
Transferring human motion to a mobile robotic manipulator and ensuring safe physical human-robot interaction are crucial steps towards automating complex manipulation tasks in human-shared environments. In this work we present a robot whole-body teleoperation framework for human motion transfer. We validate our approach through several experiments using the TIAGo robot, showing this could be an easy way for a non-expert to teach a rough manipulation skill to an assistive robot.
[ Paper ]
This is pretty cool looking for an autonomous boat, but we’ll see if they can build a real one by 2020 since at the moment it’s just an average rendering.
[ ProMare ]
I had no idea that asparagus grows like this. But, sure does make it easy for a robot to harvest.
[ Inaho ]
Skip to 2:30 in this Pepper unboxing video to hear the noise it makes when tickled.
[ HIT Lab NZ ]
In this interview, Jean Paul Laumond discusses his movement from mathematics to robotics and his career contributions to the field, especially in regards to motion planning and anthropomorphic motion. Describing his involvement at CNRS and in other robotics projects, such as HILARE, he comments on the distinction in perception between the robotics approach and a mathematics one.
[ IEEE RAS History ]
Here’s a couple of videos from the CMU Robotics Institute archives, showing some of the work that took place over the last few decades.
[ CMU RI ]
In this episode of the Artificial Intelligence Podcast, Lex Fridman speaks with David Ferrucci from IBM about Watson and (you guessed it) artificial intelligence.
David Ferrucci led the team that built Watson, the IBM question-answering system that beat the top humans in the world at the game of Jeopardy. He is also the Founder, CEO, and Chief Scientist of Elemental Cognition, a company working engineer AI systems that understand the world the way people do. This conversation is part of the Artificial Intelligence podcast.
[ AI Podcast ]
This week’s CMU RI Seminar is by Pieter Abbeel from UC Berkeley, on “Deep Learning for Robotics.”
Programming robots remains notoriously difficult. Equipping robots with the ability to learn would by-pass the need for what otherwise often ends up being time-consuming task specific programming. This talk will describe recent progress in deep reinforcement learning (robots learning through their own trial and error), in apprenticeship learning (robots learning from observing people), and in meta-learning for action (robots learning to learn). This work has led to new robotic capabilities in manipulation, locomotion, and flight, with the same approach underlying advances in each of these domains.
[ CMU RI ] Continue reading
After 25 million games, the AI agents playing hide-and-seek with each other had mastered four basic game strategies. The researchers expected that part.
After a total of 380 million games, the AI players developed strategies that the researchers didn’t know were possible in the game environment—which the researchers had themselves created. That was the part that surprised the team at OpenAI, a research company based in San Francisco.
The AI players learned everything via a machine learning technique known as reinforcement learning. In this learning method, AI agents start out by taking random actions. Sometimes those random actions produce desired results, which earn them rewards. Via trial-and-error on a massive scale, they can learn sophisticated strategies.
In the context of games, this process can be abetted by having the AI play against another version of itself, ensuring that the opponents will be evenly matched. It also locks the AI into a process of one-upmanship, where any new strategy that emerges forces the opponent to search for a countermeasure. Over time, this “self-play” amounted to what the researchers call an “auto-curriculum.”
According to OpenAI researcher Igor Mordatch, this experiment shows that self-play “is enough for the agents to learn surprising behaviors on their own—it’s like children playing with each other.”
Reinforcement is a hot field of AI research right now. OpenAI’s researchers used the technique when they trained a team of bots to play the video game Dota 2, which squashed a world-champion human team last April. The Alphabet subsidiary DeepMind has used it to triumph in the ancient board game Go and the video game StarCraft.
Aniruddha Kembhavi, a researcher at the Allen Institute for Artificial Intelligence (AI2) in Seattle, says games such as hide-and-seek offer a good way for AI agents to learn “foundational skills.” He worked on a team that taught their AllenAI to play Pictionary with humans, viewing the gameplay as a way for the AI to work on common sense reasoning and communication. “We are, however, quite far away from being able to translate these preliminary findings in highly simplified environments into the real world,” says Kembhavi.
AI agents construct a fort during a hide-and-seek game developed by OpenAI.
In OpenAI’s game of hide-and-seek, both the hiders and the seekers received a reward only if they won the game, leaving the AI players to develop their own strategies. Within a simple 3D environment containing walls, blocks, and ramps, the players first learned to run around and chase each other (strategy 1). The hiders next learned to move the blocks around to build forts (2), and then the seekers learned to move the ramps (3), enabling them to jump inside the forts. Then the hiders learned to move all the ramps into their forts before the seekers could use them (4).
The two strategies that surprised the researchers came next. First the seekers learned that they could jump onto a box and “surf” it over to a fort (5), allowing them to jump in—a maneuver that the researchers hadn’t realized was physically possible in the game environment. So as a final countermeasure, the hiders learned to lock all the boxes into place (6) so they weren’t available for use as surfboards.
An AI agent uses a nearby box to surf its way into a competitor’s fort.
In this circumstance, having AI agents behave in an unexpected way wasn’t a problem: They found different paths to their rewards, but didn’t cause any trouble. However, you can imagine situations in which the outcome would be rather serious. Robots acting in the real world could do real damage. And then there’s Nick Bostrom’s famous example of a paper clip factory run by an AI, whose goal is to make as many paper clips as possible. As Bostrom told IEEE Spectrum back in 2014, the AI might realize that “human bodies consist of atoms, and those atoms could be used to make some very nice paper clips.”
Bowen Baker, another member of the OpenAI research team, notes that it’s hard to predict all the ways an AI agent will act inside an environment—even a simple one. “Building these environments is hard,” he says. “The agents will come up with these unexpected behaviors, which will be a safety problem down the road when you put them in more complex environments.”
AI researcher Katja Hofmann at Microsoft Research Cambridge, in England, has seen a lot of gameplay by AI agents: She started a competition that uses Minecraft as the playing field. She says the emergent behavior seen in this game, and in prior experiments by other researchers, shows that games can be a useful for studies of safe and responsible AI.
“I find demonstrations like this, in games and game-like settings, a great way to explore the capabilities and limitations of existing approaches in a safe environment,” says Hofmann. “Results like these will help us develop a better understanding on how to validate and debug reinforcement learning systems–a crucial step on the path towards real-world applications.”
Baker says there’s also a hopeful takeaway from the surprises in the hide-and-seek experiment. “If you put these agents into a rich enough environment they will find strategies that we never knew were possible,” he says. “Maybe they can solve problems that we can’t imagine solutions to.” Continue reading