Abstract
Legged robots have the potential to traverse complex terrain and access confined spaces beyond the reach of traditional platforms thanks to their ability to carefully select footholds and flexibly adapt their body posture while walking. However, robust deployment in real-world applications is still an open challenge. In this paper, we present a method for legged locomotion control using reinforcement learning and 3D volumetric representations to enable robust and versatile locomotion in confined and unstructured environments. By employing a two-layer hierarchical policy structure, we exploit the capabilities of a highly robust low-level policy to follow 6D commands and a high-level policy to enable three-dimensional spatial awareness for navigating under overhanging obstacles. Our study includes the development of a procedural terrain generator to create diverse training environments. We present a series of experimental evaluations in both simulation and real-world settings, demonstrating the effectiveness of our approach in controlling a quadruped robot in confined, rough terrain. By achieving this, our work extends the applicability of legged robots to a broader range of scenarios.
Abstract (translated)
腿式机器人通过谨慎选择立足点并灵活适应身体姿势，在行走过程中具有穿越复杂地形和访问传统平台无法触及的空间的潜力。然而，在现实应用中，腿式机器人的稳健部署仍然是一个尚未解决的问题。在本文中，我们提出了使用强化学习和3D体积表示来控制腿式移动的方法，以实现如何在局限和无结构环境中进行稳健和多功能的移动。通过采用双层层次策略结构，我们利用具有高度稳健的低级别策略来遵循6D指令，并利用高级策略实现对下落障碍物的三维空间感知。我们的研究包括开发了一个程序化地形生成器，以创建多样化的训练环境。我们在仿真和现实环境中进行了系列实验评估，证明了我们在控制陷入、崎岖不平的地形中行走的四足机器人方面的有效方法。通过实现这一目标，我们的工作将腿式机器人的应用范围扩展到了更广泛的场景。
URL
https://arxiv.org/abs/2403.00187
