Neural Assistive Impulses: Synthesizing Exaggerated Motions for Physics-based Characters

2026-04-06

AI Summary

This work addresses the challenge of stably generating physically implausible yet expressive character motions—such as instantaneous sprints or abrupt mid-air trajectory changes—within physics-driven animation frameworks, where conventional external force control often leads to training instability. The authors propose an auxiliary impulse-based neural control framework that reformulates auxiliary signals in impulse space for the first time, thereby avoiding high-magnitude force spikes caused by velocity discontinuities. By decomposing motion into high-frequency components resolved through inverse dynamics and low-frequency residuals learned by a neural network, the method enables stable tracking of dynamically infeasible actions. This approach significantly enhances the robustness and expressiveness of physics-driven characters when performing highly agile maneuvers.

Abstract

Physics-based character animation has become a fundamental approach for synthesizing realistic, physically plausible motions. While current data-driven deep reinforcement learning (DRL) methods can synthesize complex skills, they struggle to reproduce exaggerated, stylized motions, such as instantaneous dashes or mid-air trajectory changes, which are required in animation but violate standard physical laws. The primary limitation stems from modeling the character as an underactuated floating-base system, in which internal joint torques and momentum conservation strictly govern motion. Direct attempts to enforce such motions via external wrenches often lead to training instability, as velocity discontinuities produce sparse, high-magnitude force spikes that prevent policy convergence. We propose Assistive Impulse Neural Control, a framework that reformulates external assistance in impulse space rather than force space to ensure numerical stability. We decompose the assistive signal into an analytic high-frequency component derived from Inverse Dynamics and a learned low-frequency residual correction, governed by a hybrid neural policy. We demonstrate that our method enables robust tracking of highly agile, dynamically infeasible maneuvers that were previously intractable for physics-based methods.

Problem

• physics-based animation
• exaggerated motion
• underactuated system
• training instability
• dynamically infeasible maneuvers

Innovation

• impulse-based control
• physics-based animation
• neural character control
• exaggerated motion synthesis
• hybrid neural policy