QuadKAN: KAN-Enhanced Quadruped Motion Control via End-to-End Reinforcement Learning
This work addresses robust locomotion for quadruped robots in complex environments, representing an incremental improvement with a novel method for a known bottleneck.
The paper tackled vision-guided quadruped motion control by proposing QuadKAN, a spline-parameterized policy using Kolmogorov-Arnold Networks, which improved sample efficiency, reduced action jitter and energy consumption, and achieved higher returns, greater distances, and fewer collisions than state-of-the-art baselines across diverse terrains.
We address vision-guided quadruped motion control with reinforcement learning (RL) and highlight the necessity of combining proprioception with vision for robust control. We propose QuadKAN, a spline-parameterized cross-modal policy instantiated with Kolmogorov-Arnold Networks (KANs). The framework incorporates a spline encoder for proprioception and a spline fusion head for proprioception-vision inputs. This structured function class aligns the state-to-action mapping with the piecewise-smooth nature of gait, improving sample efficiency, reducing action jitter and energy consumption, and providing interpretable posture-action sensitivities. We adopt Multi-Modal Delay Randomization (MMDR) and perform end-to-end training with Proximal Policy Optimization (PPO). Evaluations across diverse terrains, including both even and uneven surfaces and scenarios with static or dynamic obstacles, demonstrate that QuadKAN achieves consistently higher returns, greater distances, and fewer collisions than state-of-the-art (SOTA) baselines. These results show that spline-parameterized policies offer a simple, effective, and interpretable alternative for robust vision-guided locomotion. A repository will be made available upon acceptance.