A Tactile Feedback Approach to Path Recovery after High-Speed Impacts for Collision-Resilient Drones

TL;DR

This work addresses the vulnerability of small aerial robots to high-speed collisions in cluttered environments by leveraging lightweight, binary tactile sensors mounted on an icosahedral tensegrity drone. It combines a collision-aware state estimator that uses pre-collision velocities and contact data to better predict post-collision dynamics with a vector-field path representation that guarantees convergence to a user-defined trajectory while incorporating repulsive potentials from known obstacles to avoid re-collisions. A reflexive recovery mechanism provides an immediate stabilization maneuver, followed by adaptive path replanning that updates the obstacle representation and guides the drone back to its original path. Validation via Monte Carlo simulations and real-world experiments demonstrates robust collision recovery, accurate state estimation during impact, and successful path adjustment at high speeds, illustrating a lightweight, low-computation approach to collision resilience suitable for small drones.

Abstract

Aerial robots are a well-established solution for exploration, monitoring, and inspection, thanks to their superior maneuverability and agility. However, in many environments, they risk crashing and sustaining damage after collisions. Traditional methods focus on avoiding obstacles entirely, but these approaches can be limiting, particularly in cluttered spaces or on weight-and compute-constrained platforms such as drones. This paper presents a novel approach to enhance drone robustness and autonomy by developing a path recovery and adjustment method for a high-speed collision-resilient aerial robot equipped with lightweight, distributed tactile sensors. The proposed system explicitly models collisions using pre-collision velocities, rates and tactile feedback to predict post-collision dynamics, improving state estimation accuracy. Additionally, we introduce a computationally efficient vector-field-based path representation that guarantees convergence to a user-specified path, while naturally avoiding known obstacles. Post-collision, contact point locations are incorporated into the vector field as a repulsive potential, enabling the drone to avoid obstacles while naturally returning to its path. The effectiveness of this method is validated through Monte Carlo simulations and demonstrated on a physical prototype, showing successful path following, collision recovery, and adjustment at speeds up to 3.7 m/s.

Figures (2)

• Figure S1: A path (black dashed line) is represented by a vector field (gray arrows) that guarantees convergence from any location while avoiding known obstacles (top left). Using the pre-collision velocity ($\bm{v}^-$ and $\bm{\omega}^-$) and contact points on the drone frame, we predict the post-collision velocity ($\bm{v}^+$ and $\bm{\omega}^+$) and updated state estimate to facilitate recovery (top right). Together, these enable the to withstand collisions and adapt its path to prevent future ones (bottom).
• Figure S2: A free-body diagram of the forces and torques acting on the as well as the frames used to model it colliding with the environment. The tensegrity drone with its twelve nodes ($v_1$ through $v_{12}$) is controlled through its control forces and torques ($f_1$ and $\tau_1$ through $f_4$ and $\tau_4$) and experiences the collision impulses $\bm{j}_i$ at the $i$-th vertex.