The hottest MIT R & D blind robot can freely pass

  • Detail

MIT R & D blind robot can freely pass through dark dangerous areas

figure: the third generation hybrid flex 0.2 ⑷/4 variety two-component metering machine equipped with servo electric metering plunger is used for abrasive component a cheetah robot, which can jump on a table nearly 30 inches high

this article is produced by Yi intelligent Studio (official account smartman163). Focus on AI and understand the next era

[Yi intelligent news on July 8] robot, do you want it or not? This special series explores the evolving relationship between humans and machines, and studies the impact of robots, artificial intelligence and automation on our work and life

the intelligent robot of Boston Dynamics uses complex computer vision, but the third generation cheetah robot of MIT adopts a different strategy

the third generation of vision free cheetah robot developed by MIT weighs 80 pounds and is about the size of a Labrador dog. It can walk through a dark room or climb an obstacle stair without using a camera or environmental sensor. It relies on what engineers call "blind locomotion" - that is, the feedback of its robot legs, and the algorithm based sense of balance it needs when walking through the dark divided into a microcomputer controlled four ball experimental machine and a digital four ball experimental machine

sangbae Kim, an associate professor of mechanical engineering at MIT, said at today's press conference on the use of computer network technology: "robots should be able to handle many unexpected behaviors without relying too much on visual systems."

kim said, "vision may be noisy, inaccurate, and sometimes even completely unavailable. If it depends too much on vision, the robot must locate very accurately, which will cause it to move very slowly. Therefore, we hope that the robot will rely more on tactile information. In this way, it can handle unexpected obstacles while moving fast."

this strategy is very suitable for robots to carry out activities in disaster areas or other dangerous environments

Kim said, "the design purpose of the third generation cheetah robot is to do all kinds of tasks, which involve all kinds of terrain conditions, including stairs, curbs and the ground full of obstacles. I think in many cases, we will want robots to replace humans to complete simple tasks. Through remote control of robots, 1. In experimental operations, we can complete dangerous, dirty and difficult tasks more safely."

The team of

kim developed two new algorithms for the cheetah robot without vision system

the contact detection algorithm helps the robot determine the best time point for the transition between leg swing and stepping on the ground, which depends on the obstacles it feels when touching the ground. The algorithm calculates these probabilities based on data from gyroscopes, accelerometers, and leg joint positions

Kim explained, "if humans close their eyes and take a step forward, we will have a mental model of where the ground may be, and we can be prepared for it. But we also rely on the feeling of touching the ground. For the third generation cheetah robot, we do something similar, combining information from multiple sources to determine the switching time."

researchers tested the algorithm in experiments: let the third generation cheetah robot trot on the treadmill in the laboratory and climb stairs. Some objects, such as wood blocks and tape rolls, are scattered on both surfaces

at the same time, a model predictive control algorithm is that when any leg contacts the ground and applies a specific force, the model predictive control algorithm will immediately calculate the position of the robot's body and legs in the next half second

kim said, "for example, if someone kicks the robot from the side, when the robot's foot has touched the ground, the algorithm will decide 'how can I determine the force to be applied by the foot? Because there is a force on the left that hinders my speed, I have to apply a force in the opposite direction to eliminate that speed. If I apply a 100 Newton force in the opposite direction, what will happen in half a second?'"

the prediction algorithm calculates 20 times per second. In order to test its performance, in the experiment, researchers kicked and pulled the robot with force when the robot was on the treadmill or upstairs, and then adjusted the algorithm accordingly. (we hope that the third generation cheetah robot will not bear a grudge against this.)

finally, Kim and his colleagues will add computer vision system to the existing foundation, but for the third generation cheetah robot, the first thing they want to do is blind eye movement

a robot that can walk, run or climb in the dark, and its speed is certainly much faster than human beings? This is exactly what you want to see after the earthquake - and this is exactly what you don't want to see after the Robot Uprising

at the International Conference on intelligent robots and Systems held in October this year, non vision technology and other improvements to the cheetah robot model will become the theme of this conference. The research was supported by naver, Toyota research, Foxconn and the U.S. Air Force Office of scientific research

(from: geekwire compilation: Yi intelligent participation: Rosie)

pay attention to Yi intelligent official account (smartman163), and interpret the big events of large companies in the AI field, new ideas and new applications for you

Copyright © 2011 JIN SHI