LogInFrom a person to a chair to a robot.
If you thought Souta from Suzume had a hard life, wait until you see researchers from JSK Robotics Laboratory at The University of Tokyo slapping the poor chair to make it stand up over and over again.
The scientists recreated the three-legged chair as a robot and programmed it to walk and find its footing if it fell.
"Its body structure consists of three legs that are asymmetric to the body, so it cannot be easily balanced. In addition, the actuator is a servo motor that can only feed-forward rotational angle commands and the sensor can only sense the robot's posture quaternion."
The robot has three gimbal-type 2-DOF legs rotated by two servo motors.
Image credit: Shintaro Inoue et al.
Image credit: Shintaro Inoue et al.
The authors generated gaits with two different methods: using linear completion to connect the postures necessary for the gait discovered through trial and error using the robot and using the gait generated by reinforcement learning in the simulator and reflecting it to the actual robot.
Reinforcement learning is a machine learning method that teaches an agent to do tasks in such a way that it can maximize the reward.
Image credit: Shintaro Inoue et al.
Image credit: Shintaro Inoue et al.
The researchers concluded that both methods can make the robot walk and stand up, although they lead to slightly different results as you can see in the video.
Which method works better in your opinion? Find the project and its code here and join our 80 Level Talent platform and our Telegram channel, follow us on Instagram, Twitter, and LinkedIn, where we share breakdowns, the latest news, awesome artworks, and more.