Lots to giggle at falling robot. Researchers at the University of Lorraine have develop a “Damage Reflex” (aka D-Reflex) system that has the humanoid robot TALOS prop itself against a wall when one of its legs is broken, like a human having just lost its balance. The neural network-based system uses its experience (in this case, 882,000 training simulations) to quickly find a point on the wall that is most likely to provide stability. The robot doesn’t care how damaged it is and can approach as quickly as a human.
Result, like IEEE Phổ Spectrum Note, is the counter-comedy you’d expect. Instead of tumbling to the ground, the robot braced itself against the wall like someone who had just sprained their ankle. It wasn’t particularly graceful and required the robot to stop on contact, but it was effective in three out of four tests.
D-Reflex is not guaranteed to prevent falls, partly because it cannot account for every possible location or surface. It also doesn’t help the robot recover from a disaster — you won’t see the robot limp along the wall until it finds help. The current approach is also based on a stationary bot and won’t help if the actuator fails midway.
However, the researchers hope to create a useful system on the go, and envision robots that can grab chairs and other complex objects when they’re about to fall. This can save replacement cost worker robot otherwise that would lead to their demise and possibly lead to more ‘natural’ bots learning to use their environment to their advantage. One thing is for sure: if robot happen, tripping won’t stop them for long.
All products recommended by Engadget are selected by our editorial team, independent of our parent company. Some of our stories include affiliate links. If you buy something through one of these links, we may earn an affiliate commission. All prices are correct at time of publication.