Robotics

New UAV Control Framework Ensures Stable Contact During Aerial Physical Interaction

Researchers redefine full actuation for drones that push, pull, and sustain contact.

Deep Dive

Researchers Abhimanyu Khadga, Abhinav Sinha, and Shashi Ranjan Kumar introduce a control-theoretic framework for contact-persistent full actuation in unmanned aerial vehicles (UAVs). Traditional certification of fully actuated UAVs relies on rank conditions of the control-allocation matrix or free-flight tracking performance, which is insufficient for aerial physical interaction tasks. During sustained contact, part of the available wrench is consumed by the interaction task, leaving only a residual wrench for stabilization and maneuvering. The paper defines feasible wrench sets under actuator limits, residual wrench sets under task loading, and residual authority margins that strengthen the usual rank-based notion of full actuation.

The main theoretical result shows that contact-persistent full actuation is equivalent to the interiority of the task wrench within the constrained feasible wrench polytope, with the residual authority radius being exactly the distance to the polytope boundary. The framework also introduces a signed residual-margin certificate for infeasible/boundary cases and a slack-maximizing allocation certificate. Numerical evaluation on an abstract tilted hexarotor demonstrates that full row rank alone does not imply feasible contact operation. Intermediate tilt angles preserve residual authority during pushing, while small or excessive tilts fail due to lateral-force deficiency or hover-margin loss. This work provides a rigorous tool for designing UAVs capable of sustained physical interaction.

Key Points
  • Replaces rank-based full actuation certification with a polytope interiority condition that accounts for task wrench consumption.
  • Residual authority radius equals the distance from the task wrench to the boundary of the feasible wrench polytope.
  • Numerical evaluation on a tilted hexarotor shows intermediate tilt angles (~30-60°) preserve authority, while small (<20°) or large (>70°) tilts fail.

Why It Matters

Enables drones to reliably perform sustained physical tasks like pushing, lifting, or manipulating objects in real-world environments.

📬 Get the top 10 AI stories daily