Learning Humanoid Locomotion with World Model Reconstruction
This addresses the gap in developing controllers for humanoid robots to operate in uncontrolled, complex environments, representing a strong specific gain rather than a broad paradigm shift.
The study tackled the problem of enabling humanoid robots to navigate complex real-world terrains by introducing World Model Reconstruction (WMR), an end-to-end learning-based approach that reconstructs world states to enhance locomotion policies, resulting in the robot successfully completing a 3.2 km hike on challenging surfaces like ice and snow without human assistance.
Humanoid robots are designed to navigate environments accessible to humans using their legs. However, classical research has primarily focused on controlled laboratory settings, resulting in a gap in developing controllers for navigating complex real-world terrains. This challenge mainly arises from the limitations and noise in sensor data, which hinder the robot's understanding of itself and the environment. In this study, we introduce World Model Reconstruction (WMR), an end-to-end learning-based approach for blind humanoid locomotion across challenging terrains. We propose training an estimator to explicitly reconstruct the world state and utilize it to enhance the locomotion policy. The locomotion policy takes inputs entirely from the reconstructed information. The policy and the estimator are trained jointly; however, the gradient between them is intentionally cut off. This ensures that the estimator focuses solely on world reconstruction, independent of the locomotion policy's updates. We evaluated our model on rough, deformable, and slippery surfaces in real-world scenarios, demonstrating robust adaptability and resistance to interference. The robot successfully completed a 3.2 km hike without any human assistance, mastering terrains covered with ice and snow.