AI News, Video Friday: AI vs. Dota 2, Cassie Gets Bored, and Georgia Tech's Robotarium
- On Tuesday, February 13, 2018
- By Read More
Video Friday: AI vs. Dota 2, Cassie Gets Bored, and Georgia Tech's Robotarium
Video Friday is your weekly selection of awesome robotics videos, collected by your Automaton bloggers.
won't pretend to know what Dota stands for, or what happened to Dota 1 (something bad I assume?), but I guess Dota 2 is a big deal, because OpenAI seem pretty stoked that the programmed an AI that can beat the best human players in a 1v1:
Alex, who has a fresh Ph.D in robotics from CMU and also spent some time at HEBI Robotics, now works for a company that supports robotics projects for certain government agencies with bunches of letters for names.
In addition to these 25 parts, 16 servo motors, 3 motor controllers, an Arduino Due and a number of other components were necessary to fully assemble the robot.
Will wrote in to share this video from Sony, which (as he points out) is interesting because it was published just a few weeks ago but features AIBO and QRIO as sort of the flagship expression of the current (public) state of Sony's robotics and AI.
It's entirely possible that an AIBO ERS-7 from 2005 would give the forthcoming generation of social home robots a run for their money in terms of overall capabilities and interactivity, which is bananas, considering how long ago 2005 is in robotics years.
And this is why Cassies do this, according to Agility Robotics: The 'looking around' behavior was added for safety reasons: the balancing control loop runs so quickly that the robot appears stationary to many observers.
It's a bit like personal space for automobiles - you're comfortable standing up against a parked car, but stay further away if it's obvious that the engine is on.
The Free Gait interface defines a whole-body abstraction layer to accommodate a variety of task-space control commands such as end effector, joint, and base motions.
The deﬁned motion tasks are tracked with a feedback whole-body controller to ensure accurate and robust motion execution even under slip and external disturbances.
The application of this framework includes intuitive tele-operation of the robot, efficient scripting of behaviors, and fully autonomous operation with motion and footstep planners.
On August 15th WeRobotics hosted a webinar featuring Alexander Fraser entitled: “Witchcraft and Explosions: how do community perceptions to humanitarian drones differ in Tanzania and Malawi?” Alexander, who is a BSc Geography Undergraduate student at the University of Edinburgh, spoke about his recent experiences shadowing the WeRobotics team in Zanzibar and a UNICEF team in Kasungu, conducting interviews with community elders, chiefs, matriarchs, village citizens, and district officials to assess their understandings of drones, or 'ndege'.
Some of the goals of the research were to investigate which drone applications would benefit the local communities, what are their fears regarding drone use, and to identify sites where flying a drone would be considered inappropriate.
How to build an autonomous, voice-controlled, face-recognizing drone for $200
I got my feet wet first by building a drone from parts, but like pretty much all of my DIY projects, building from scratch ended up costing me way more than buying a prebuilt version—and frankly, my homemade drone never flew quite right.
You can buy one for $200 new, but so many people buy drones and never end up using them that a secondhand drone is a good option and available widely on eBay for $130 or less.
The Parrot AR drone doesn’t fly quite as stably as the much more expensive (about $550) new Parrot Bebop 2 drone, but the Parrot AR comes with an excellent node.js client library called node-ar-drone that is perfect for building onto.
This setup led to lower latency than running a neural network directly on Raspberry PI hardware, and I think this architecture makes sense for hobby drone projects at the moment.
After installing the node library, it’s fun to make a node.js REPL (Read-Evaluate-Print-Loop) and steer your drone: If you are actually following along, by now you’ve definitely crashed your drone—at least a few times.
found the best way to send a feed from the drone’s camera was to open up a connection and send a continuous stream of PNGs for my webserver to my website. My webserver continuously pulls PNGs from the drone’s camera using the AR drone library.
The trickiest part about doing speech recognition was not the speech recognition itself, but streaming audio from a webpage to my local server in the format Microsoft’s Speech API wants, so that ends up being the bulk of the code.
After crashing my drone into the furniture and houseplants one too many times in my livingroom, my wife nicely suggested I move my project to my garage, where there is less to break—but there isn’t much room to maneuver (see Figure 3).
When I get a bigger lab space, I’ll work more on smart searching algorithms, but for now I’ll just have my drone take off and rotate, looking for my friends and enemies: Putting it all together Check out this video I took of my drone taking off and searching for my friend Chris: Once everything is set up and you are controlling the drone through an API and getting the video feed, hacking on drones becomes incredibly fun.
At first, I was worried that the drone’s unusually wide-angle camera might affect the face recognition and that the loud drone propeller might interfere with the speech recognition, but overall the performance was much better than I expected.
Lethal Microdrones, Dystopian Futures, and the Autonomous Weapons Debate
The video, called “Slaughterbots” and produced with support from Elon Musk’s Future of Life Institute, combines graphic violence with just enough technical plausibility to imagine a very bleak scenario: A fictional near future in which autonomous explosive-carrying microdrones are killing thousands of people around the world.
The message here seems to be that if you’re interested in having a discussion about these issues, or if you think that perhaps there might be other, potentially more effective ways of shaping the future of autonomous weapons besides calling for a ban, then you’re siding with terrorists.
Saying “we have an opportunity to prevent the future you just saw,” as UC Berkeley professor Stuart Russell, one of the creators of“Slaughterbots,” does at the end of the video, is an oversimplification—in my opinion, a ban on autonomous weapons won’t prevent the miniaturization of drones, won’t prevent advances in facial-recognition technology, and likely won’t prevent the integration of the two, which is just the scenario that “Slaughterbots” presents.
It shows what happens if you put together some existing technologies and the misuses to which such technology could be put.” Two years ago, I responded to the first open letter calling for a ban on offensive autonomous weapons by presenting an alternate perspective about why autonomous weapons might also be beneficial.
Having this discussion now is the best way to work towards a more peaceful future, but to do that, we have to have the discussion, not make scary videos designed to vilify the people we disagree with while reinforcing the unrealistic and increasingly negative perception that many people already have about robotics.
- On Saturday, January 19, 2019
The Future of Flying Robots | Vijay Kumar | TED Talks
At his lab at the University of Pennsylvania, Vijay Kumar and his team have created autonomous aerial robots inspired by honeybees. Their latest breakthrough: Precision Farming, in which swarms...
If this isn't what you want, please take action at
Robots that fly ... and cooperate | Vijay Kumar
In his lab at Penn, Vijay Kumar and his team build flying quadrotors, small, agile robots that swarm, sense each other, and form ad hoc teams -- for construction, surveying..
THE FUTURE OF MILITARY AI ROBOTICS AND COMBAT DRONES BUILT FOR WAR 2017!
The AI (Artificial Intelligence) Revolution is here and has a lot of people very worried about the future AI and robots merging. This 2017 one hundred and sixteen roboticists and AI researchers,...
"Killer" Drones Slaughterbots - Autonomous Weapons Warning
AI Experts Make a Terrifying Film Calling for a Ban on Killer Robots The Future of Life Institute, an AI watchdog organization supported by the likes of Elon Musk and Stephen Hawking, has...
Meet the dazzling flying machines of the future | Raffaello D'Andrea
When you hear the word "drone," you probably think of something either very useful or very scary. But could they have aesthetic value? Autonomous systems expert Raffaello D'Andrea develops...
What do advancements in AI mean for the military? Military robotics has come a long way with advancements in machine learning, the soaring affordability of computing power, and the rise of...
Robots and drones
A rumbling robot and several high-flying drones recently made an on-site appearance at Clemson University to burrow through and buzz above 15 acres of experimental sorghum plots containing...
Arduino DS WiFi Camera Robot - Assembly and Presentation
Arduino Playground: Windows and Android software here: Arduino DS WiFi Robot Kit with Real-time Video function! You can buy the kit from GearBest,.
Air Force Bugbot Nano Drone Technology
Air Force Bugbots Nano Drone video gives a peak inside what nano-drone technology the Federal Government is currently implementing within the united states more than a scary thought or sci-fi...