A Closed-Form CLF-CBF Controller for Whole-Body Continuum Soft Robot Collision Avoidance

A team from MIT, including Kiwan Wong and Daniela Rus, has published a breakthrough paper on arXiv detailing a new control algorithm for soft robots. The research tackles a critical challenge: enabling safe, real-time operation for soft continuum manipulators in cluttered, 3D environments. While these robots are inherently safer due to mechanical compliance, existing control methods for collision avoidance, like sampling-based planning or optimization-based Control Lyapunov Function–Control Barrier Function (CLF-CBF) approaches, are too computationally slow and lack formal safety guarantees for reliable deployment.

The team's innovation is a 'closed-form' CLF-CBF controller. Instead of solving a complex optimization problem in real-time (a quadratic program), their method analytically derives a control input that inherently satisfies safety constraints. This eliminates the computational bottleneck and potential 'feasibility' issues of online solvers. The result is dramatic speed gains—the controller is up to 10 times faster than standard CLF-CBF quadratic-programming methods and a staggering 100 times faster than traditional sampling-based planners. This performance leap was validated in both simulation and hardware experiments on a tendon-driven soft manipulator, demonstrating accurate trajectory tracking and robust obstacle avoidance.

This work represents a significant step toward practical, safe soft robotics. By providing a scalable and provably safe control strategy that operates in real-time, it unlocks the potential for soft robots to be deployed in dynamic, safety-critical settings like healthcare, search-and-rescue, or advanced manufacturing. The framework ensures stability and safety under its modeling assumptions, moving the field beyond heuristic or computationally prohibitive methods toward formally verified, high-performance control.