Constrained MPC-Based Motion Planning for Morphing Quadrotors in Ultra-Narrow Passages under Limited Perception

Researchers proposed a nonlinear MPC method for morphing quadrotors to pass through ultra-narrow gaps using only 2D LiDAR sensing.

The framework combines a smooth exponential obstacle cost with a cost-reduction factor, enabling joint planning of vehicle morphology and trajectory.

Simulations and real-world tests showed successful traversal in corridors where conventional obstacle-avoidance methods failed.

The approach could improve safe navigation in cluttered, tight spaces and may generalize to other mobile robots.