Cassie ran quarter miles on different terrains while performing long and high jumps, without being trained on each movement
Thanks to a new method based on the Al technique named reinforcement learning, robots can learn to deal with unpredictable situations encountered in this way. Reinforcement learning is like pet training; it involves rewards and penalties .Cassie is a human-sized robot. He has a height of 3,2 feet and weighs 68 pounds , and his two legs got the backup, so he ran quarter miles on different terrains while performing long and high jumps without being trained on each movement.
As described, they used a simulation to train Cassie because that speeds up the time it takes to learn and enables the robot to perform those same skills in the real world.They trained the neural network that controlled Cassie to master simple skills like jumping on the spot and walking forward. They presented the model with new commands encouraging the robot to perform tasks using its new movement skills.
So, when it became good with new tasks in a simulated environment, they then used a method of task randomization to diversify the tasks he had been trained on. That could help him be more prepared for unexpected situations. As described in the study, they incorporated a task completion term into the reward to ensure the robot accomplishes the assigned tasks using the acquired locomotion skill. Within this term, they motivated the robot to align its movement with the desired velocity and turning rate…
United States Latest News, United States Headlines
Similar News:You can also read news stories similar to this one that we have collected from other news sources.
Training dairy cows with positive reinforcement can turn otherwise stressful events into playDairy cows receive a huge amount of care during their lives on a farm, but procedures, especially restraining during procedures, can be stressful for some animals. Positive reinforcement training -- or training with a reward to achieve a desired behavior -- has shown promise to reduce this fear in other species.
Read more »
Human-to-humanoid robot full-body teleoperation unlocked in real-timeResearchers unveil live reinforcement learning-based whole-body humanoid teleoperation system, promising significant advancements.
Read more »
Cassie Settles Rape, Human Trafficking Lawsuit with Diddy 'Amicably'Cassie and Diddy settle her lawsuit out of court 'amicably.'
Read more »
Everett official dodges key question about gazebo-saving compromiseAn Everett spokesperson dodged KIRO Newsradio's question about whether Mayor Cassie Franklin is willing to accept the gazebo compromise.
Read more »
Domain knowledge drives data-driven artificial intelligence in well loggingData-driven artificial intelligence, such as deep learning and reinforcement learning, possesses powerful data analysis capabilities. These techniques enable the statistical and probabilistic analysis of data, facilitating the mapping of relationships between inputs and outputs without reliance on predetermined physical assumptions.
Read more »
Hailey Bieber Slams TikTok 'Blind Items' in Statement Amid Speculation About Her and Justin BieberAlix Earle, gorgeous It Girl and natural model, hung out on Coney Island for a Cosmopolitan fashion shoot (grab the Spring 2024 issue to see her in action!) Cosmo fashion director and stylist extraordinaire, Cassie Anderson, worked her magic with Alix in some of the MOST fun spring fashion we’ve seen in a while.
Read more »