Whole-body motion planning and safety-critical control for aerial manipulation

TL;DR

This work tackles safe, dynamically feasible whole-body planning for aerial manipulators by introducing a superquadric–proxy representation for both the vehicle and obstacles. A maximum-clearance planner combines Voronoi diagrams with an equilibrium-manifold formulation to generate smooth, collision-aware trajectories, while a safety-critical controller enforces thrust limits and instantaneous collision avoidance via high-order control barrier functions. Compared to sampling-based planners and ellipsoid baselines, the proposed framework delivers faster, safer, and more geometrically faithful trajectories in cluttered environments, with successful validation on a real aerial-manipulation platform. The results demonstrate strong simulation-to-hardware consistency, highlighting practical impact for safe autonomous aerial manipulation in complex settings.

Abstract

Aerial manipulation combines the maneuverability of multirotors with the dexterity of robotic arms to perform complex tasks in cluttered spaces. Yet planning safe, dynamically feasible trajectories remains difficult due to whole-body collision avoidance and the conservativeness of common geometric abstractions such as bounding boxes or ellipsoids. We present a whole-body motion planning and safety-critical control framework for aerial manipulators built on superquadrics (SQs). Using an SQ-plus-proxy representation, we model both the vehicle and obstacles with differentiable, geometry-accurate surfaces. Leveraging this representation, we introduce a maximum-clearance planner that fuses Voronoi diagrams with an equilibrium-manifold formulation to generate smooth, collision-aware trajectories. We further design a safety-critical controller that jointly enforces thrust limits and collision avoidance via high-order control barrier functions. In simulation, our approach outperforms sampling-based planners in cluttered environments, producing faster, safer, and smoother trajectories and exceeding ellipsoid-based baselines in geometric fidelity. Actual experiments on a physical aerial-manipulation platform confirm feasibility and robustness, demonstrating consistent performance across simulation and hardware settings. The video can be found at https://youtu.be/hQYKwrWf1Ak.

Figures (9)

• Figure 1: Robot arm's actuation for collision avoidance in a cluttered environment.
• Figure 2: An aerial manipulator configured with a fully actuated multirotor and a 3-DOF robot arm with the thrust vector, $\boldsymbol{T} \in {\mathbb{R}}^{6}$, the joint angles of the robot arm, $[\theta_1;\theta_2;\theta_3]$, and the Earth-fixed and multirotor frames, $\mathscr{F}_{W}$ and $\mathscr{F}_b$.
• Figure 3: Overall diagram of planning and control of aerial manipulator.
• Figure 4: System representation using SQs and proxies. The end effector of the robot arm aligns with the Voronoi edge to ensure collision-free motion.
• Figure 5: Simulation results from the whole-body motion planner. Blue transparent polygons represent Voronoi regions $V_i$ separated by grey hyperplanes $HP_{i,j}$, computed via proxies between obstacles. The aerial manipulator and obstacles are modeled by SQs (black). The red point denotes the target; magenta indicates ${\boldsymbol{\mathbf{u}}}(s)$ during ODE integration (Eq. \ref{['eq:allODE']}); and blue shows the smooth trajectory ${\boldsymbol{\mathbf{z}}}(s)$ of the aerial manipulator.