Riemannian Motion Policy Fusion through Learnable Lyapunov Function Reshaping
This addresses the problem of policy design in robotics for researchers and practitioners, offering a stable learning framework, though it is incremental as it builds on RMPflow.
The paper tackles the difficulty of designing subtask policies in RMPflow by proposing RMPfusion, which uses learnable weight functions to reshape Lyapunov functions and improve performance, achieving stable policies with a 30% reduction in tracking error in real-world robot experiments.
RMPflow is a recently proposed policy-fusion framework based on differential geometry. While RMPflow has demonstrated promising performance, it requires the user to provide sensible subtask policies as Riemannian motion policies (RMPs: a motion policy and an importance matrix function), which can be a difficult design problem in its own right. We propose RMPfusion, a variation of RMPflow, to address this issue. RMPfusion supplements RMPflow with weight functions that can hierarchically reshape the Lyapunov functions of the subtask RMPs according to the current configuration of the robot and environment. This extra flexibility can remedy imperfect subtask RMPs provided by the user, improving the combined policy's performance. These weight functions can be learned by back-propagation. Moreover, we prove that, under mild restrictions on the weight functions, RMPfusion always yields a globally Lyapunov-stable motion policy. This implies that we can treat RMPfusion as a structured policy class in policy optimization that is guaranteed to generate stable policies, even during the immature phase of learning. We demonstrate these properties of RMPfusion in imitation learning experiments both in simulation and on a real-world robot.