2024-07-15 00:52:00
AGI – A humanoid robot that is trained to seamlessly learn and execute a variety of expressive movements, including simple patterns of dance steps and gestures such as waving, high-fiving and hugging while maintaining a stable posture across different environments is shown to have a positive effect. on interactions between humans and robots in work environments.
This is what came out of a study by engineers from the University of California San Diego, presented at the conference “Robotics: Science and Systems”, which took place from 15 to 19 July in Delft, the Netherlands. The great sensitivity and power of this humanoid robot paves the way for improved human-robot interactions in environments such as industrial assembly lines, hospitals and homes, where robots can work safely alongside humans or even replace them in the hazardous areas such as laboratories. “Through expression and more human body movements, we aim to build trust and demonstrate the ability of robots to be compatible with people,” Xiaolong Wang, a professor in the Department of Electrical and Computer Science at Home -Jacobs book of UC San said. Diego Engineering.
“We are working to help reshape the public’s perception of robots as friendly and cooperative rather than terrifying like the Terminator,” Wang continued. What makes this humanoid robot so defining is that it has been trained on many human body movements, allowing it to synthesize new movements and imitate them with ease. As a fast-paced dance student, the robot can quickly learn new techniques and gestures To train the robot, the research team used a large collection of motion capture data and dance videos.
Their method is the training of the upper and lower body separately. This method allows the upper part of the robot to perform various reference movements, such as dancing and high-fiving, while the legs focus on a constant step movement to maintain balance and cross different areas.
“The main goal is to demonstrate the robot’s ability to do different things while walking from one place to another without falling,” Wang explained. Despite training the upper and lower body separately, the robot operates under a unified policy that controls its entire system. This integration policy ensures that the robot can perform complex body gestures while walking stably on surfaces such as gravel, dirt, wood chips, grass and concrete paths.
The simulations were first performed on a virtual humanoid robot and then transferred to a real robot. The robot demonstrated the ability to perform both learned and new movements in real-world situations. Currently, the robot’s movements are guided by a human operator through a game controller, which dictates its speed, direction and specific movements. The team plans a future version equipped with a camera to allow the robot to perform tasks and navigate the ground completely autonomously. A group of scientists are busy refining the robot design to tackle more demanding and demanding tasks. “By improving the upper body’s capabilities, we can expand the range of motions and gestures the robot can perform,” Wang concluded.
#Dancing #robots #work #people