PaCo-VLA: Passivity-Shielded Compliance Prior for Contact-Rich Vision-Language-Action Manipulation
This work addresses the safety-reliability gap in deploying Vision-Language-Action models for force-sensitive robotic manipulation, providing a provably safe runtime interface.
PaCo-VLA introduces a passivity-shielded compliance prior that decouples high-level semantic reasoning from low-level contact dynamics, achieving zero passivity violations and superior precision in contact-rich manipulation tasks compared to unshielded VLA baselines.
Contact-rich manipulation demands both high-level semantic reasoning and the safe regulation of high-frequency contact dynamics. While Vision-Language-Action (VLA) models provide unprecedented semantic generalization, their low-rate outputs lack the reliability required for direct plant authority in force-sensitive tasks. To bridge this semantic-to-control gap, we introduce PaCo-VLA, a passivity-shielded compliance prior that recasts the VLA interface. Rather than trusting VLAs with direct motor commands, PaCo-VLA treats network outputs as task-level compliance proposals: semantic bindings, task stages, and admittance schedules. A high-frequency, proposal-independent passivity shield governs these proposals through energy-tank accounting and boundary checks, preventing invalid, stale, or unverified model predictions from bypassing low-level contact physics. This decoupled architecture also enables causal evaluation, isolating semantic contributions from geometric shortcuts. Extensive simulated and real-world connector-insertion experiments demonstrate that PaCo-VLA achieves superior precision over unshielded VLA baselines, sustaining zero passivity violations even under adversarial compliance shifts. This framework establishes a provably sampled-passive runtime contract at the admittance port and provides a runtime interface for deploying foundation models in contact-rich domains.