Moving On, Even When You're Broken: Fail-Active Trajectory Generation via Diffusion Policies Conditioned on Embodiment and Task

2026-02-02

AI Summary

This work addresses the challenge of robotic task continuity following actuator failures, which often compromises safe operation and necessitates human intervention. The authors propose DEFT, a novel approach that introduces conditional diffusion models into fault-tolerant control for the first time. DEFT actively generates feasible trajectories conditioned on the robot’s current state and task constraints, enabling fail-active operation under arbitrary actuator faults. The method unifies constrained and unconstrained motion planning within a single framework. In simulation, DEFT achieves twice the performance of baseline methods across thousands of joint-failure scenarios spanning multiple tasks and demonstrates strong generalization to previously unseen fault conditions. Real-world experiments on a 7-DoF manipulator successfully execute complex multi-step tasks—such as opening drawers and erasing whiteboards—significantly outperforming conventional approaches.

Technology Category

• Intelligent Robots: Manipulation
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Robot failure is detrimental and disruptive, often requiring human intervention to recover. Maintaining safe operation under impairment to achieve task completion, i.e. fail-active operation, is our target. Focusing on actuation failures, we introduce DEFT, a diffusion-based trajectory generator conditioned on the robot's current embodiment and task constraints. DEFT generalizes across failure types, supports constrained and unconstrained motions, and enables task completion under arbitrary failure. We evaluated DEFT in both simulation and real-world scenarios using a 7-DoF robotic arm. In simulation over thousands of joint-failure cases across multiple tasks, DEFT outperformed the baseline by up to 2 times. On failures unseen during training, it continued to outperform the baseline, indicating robust generalization in simulation. Further, we performed real-world evaluations on two multi-step tasks, drawer manipulation and whiteboard erasing. These experiments demonstrated DEFT succeeding on tasks where classical methods failed. Our results show that DEFT achieves fail-active manipulation across arbitrary failure configurations and real-world deployments.

Problem

• fail-active
• actuation failures
• task completion
• robot failure
• embodiment

Innovation

• fail-active
• diffusion policy
• embodiment-conditioned
• trajectory generation
• actuation failure