When quadruped robots are used in earth-like gravity conditions, attitude control is mainly achieved through reaction forces with the ground. Legged robots are also well-suited for planetary exploration. There, low-gravity conditions require novel models of locomotion.
A recent study proposes a robot that can control its orientation using only its limbs. The approach is inspired by how cats safely land back on their feet after falls. Reinforcement learning is used to teach the robot several tasks of different complexity, from reaching a given pitch angle in free float to an imitation of landing on an asteroid.
The approach solves the tasks end-to-end and without using predefined trajectories. Extensive tests on the physical robot validated two-dimensional tasks and can be scaled to three dimensions and even arbitrarily shaped terrain.
In this article, we show that learned policies can be applied to solve legged locomotion control tasks with extensive flight phases, such as those encountered in space exploration. Using an off-the-shelf deep reinforcement learning algorithm, we trained a neural network to control a jumping quadruped robot while solely using its limbs for attitude control. We present tasks of increasing complexity leading to a combination of three-dimensional (re-)orientation and landing locomotion behaviors of a quadruped robot traversing simulated low-gravity celestial bodies. We show that our approach easily generalizes across these tasks and successfully trains policies for each case. Using sim-to-real transfer, we deploy trained policies in the real world on the SpaceBok robot placed on an experimental testbed designed for two-dimensional micro-gravity experiments. The experimental results demonstrate that repetitive, controlled jumping and landing with natural agility is possible.
Research paper: Rudin, N., Kolvenbach, H., Tsounis, V., and Hutter, M., “Cat-like Jumping and Landing of Legged Robots in Low-gravity Using Deep Reinforcement Learning”, 2021. Link: https://arxiv.org/abs/2106.09357