Search papers, labs, and topics across Lattice.
This paper addresses the challenge of deploying vision-language-action (VLA) models on rigid industrial robots by incorporating velocity feedforward terms into VLA policies, which are typically omitted in standard Behavior Cloning approaches. They propose two model-agnostic methods: a time-discrete finite-difference approximation and a continuous Cubic B-Spline action space, both designed to extract velocity targets from VLAs. Experimental results on a cube-in-hole task using a fine-tuned $\pi_{0.5}$ model demonstrate that the finite-difference approach improves task execution speed, while the B-Spline approach maintains high success rates and enables smoother trajectories.
Give your robot smoother moves: this work shows how to add velocity feedforward to vision-language-action models, improving both speed and compliance for contact-rich tasks.
While vision-language-action (VLA) models have shown great promise for robot manipulation, their deployment on rigid industrial robots remains challenging due to the inherent trade-off between compliance and responsiveness. Standard Behavior Cloning (BC) approaches predict discrete poses at low frequencies, omitting the velocity and acceleration feedforward terms typically used by low-level compliant controllers. This requires to rely on high stiffness for accurate tracking, thereby sacrificing safe contact dynamics. In this paper, we demonstrate the importance of integrating velocity feedforward terms into VLA policies to resolve this trade-off. We propose two methods for extracting velocity targets from VLAs: a time-discrete finite-difference approximation that serves as a highly effective bridge for existing models, and a continuous Cubic B-Spline action space that natively yields $C^2$ continuous trajectories for high-frequency control. Crucially, both approaches are strictly model-agnostic and compatible with any standard action-chunking architecture, requiring modifications only to teleoperation, data processing, and the low-level controller. We fine-tune the $\pi_{0.5}$ model and evaluate both of our approaches on a demanding, contact-rich cube-in-hole task. Our results indicate that incorporating the velocity feedforward term via finite differences significantly improves task execution speed, while the continuous B-Spline approach maintains high overall success rates and provides a foundation for smoother higher-order derivatives without compromising compliance.
