ForceVLA2: Unleashing Hybrid Force-Position Control with Force Awareness for Contact-Rich Manipulation

Abstract

Embodied intelligence for contact-rich manipulation has predominantly relied on position control, while explicit awareness and regulation of interaction forces remain under-explored, limiting stability, precision, and robustness in real-world tasks. We propose ForceVLA2, an end-to-end vision-language-action framework that equips robots with hybrid force-position control and explicit force awareness. ForceVLA2 introduces force-based prompts into the VLM expert to construct force-aware task concepts across stages, and employs a Cross-Scale Mixture-of-Experts (MoE) in the action expert to adaptively fuse these concepts with real-time interaction forces for closed-loop hybrid force--position regulation. To support learning and evaluation, we construct ForceVLA2-Dataset, containing 1000 trajectories over 5 contact-rich tasks, including wiping, pressing, and assembling, with multi-view images, task prompts, proprioceptive state, and force signals. Extensive experiments show that ForceVLA2 substantially improves success rates and reliability in contact-rich manipulation, outperforming 𝝅0 and 𝝅05 by 48.0% and 35.0%, respectively, across the 5 tasks, and mitigating common failure modes such as arm overload and unstable contact, thereby actively advancing force-aware interactive physical intelligence in VLAs.

ForceVLA2 bridges the gap between high-level sense reasoning and low-level physical interaction by transforming force from a passive perceptual input into an active hybrid force-position control policy.

• Overcoming Position-Only Limits: Current VLAs predominantly rely on position control, lacking the physical common sense and force awareness required for stable, contact-rich manipulation.

• Dual-Horizon Control Architecture: Inspired by human sensorimotor systems, ForceVLA2 addresses the need for a dual-level architecture that combines long-horizon strategic reasoning with short-horizon reactive force regulation.

• Closing the Perception-Action Loop: By integrating force as a fundamental signal, ForceVLA2 mitigates failure modes like arm overload and contact instability, advancing the robustness of robots in real-world physical environments.

Overview of ForceVLA2

Model design