A research team from the Department of Mechanical Engineering at the Korea Institute of Advanced Technology (KAIST) has made a significant breakthrough in robotics technology. They have developed a method to control a four-legged robot that can dexterously navigate deformable terrain, such as a sandy beach, by using force modeling technology and a neural network capable of making real-time decisions.

The team applied their technology to reinforcement learning, a type of AI learning method in which an agent interacts with the environment and uses the data collected to complete a task. The trained neural network controller is expected to expand the scope of four-legged walking robots by proving its resilience in changing terrain, including the ability to move at high speed on a sandy beach and walk and turn on soft ground without losing balance.

The study, titled "Learning quadrupedal locomotion on deformable terrain," is published in Science Robotics.

To create a learnable controller capable of maintaining balance on a deforming surface, the simulator must provide a similar contact experience. The research team defined a model that predicts the force generated by ground contact from the motion dynamics of a walking body based on a ground reaction force model that takes into account the additional effect of the mass of the granular medium. This allowed them to effectively simulate the deformation of the relief.

The research team also implemented an artificial neural network framework that implicitly predicts ground characteristics using a recurrent neural network that analyzes time series data from the robot's sensors.

The created controller was installed on the RaiBo robot, built by the research team, and demonstrated impressive results. RaiBo was able to walk at speeds of up to 3.03 m/s on a sandy beach, where its feet were immersed in the sand. Even when moving on harder terrains such as turf and treadmills, RaiBo was able to run consistently, adapting to the characteristics of the ground without additional programming or overhaul of the control algorithm.

This modeling and learning methodology developed by the research team is expected to facilitate the performance of practical tasks by robots, paving the way for new opportunities in robot jobs and robot worker applications in various industries.

Hire a robot equipped with this advanced technology to navigate challenging terrains with ease and efficiency, opening new possibilities for robotics in fields such as search and rescue, exploration, and more.