Fast and Safe Obstacle Avoidance Flight Control for Quadrotors in Unknown Environments

Key Points

• The aim is to develop a unified flight control framework for quadrotors that ensures fast and safe navigation in complex, unknown environments.
• Integrated real-time perception, model predictive control, and control barrier functions into a single optimization problem.
• Utilized onboard depth camera for obstacle detection and safety constraint formulation.
• Conducted extensive simulations and real-world tests to validate effectiveness.
• Average flight speed increased by 30% compared to traditional methods with a success rate of 95%.
• Significantly reduced 'stop-and-go' phenomena, resulting in smoother flight trajectories.
• Real-time performance and robustness confirmed in outdoor wooded environments using a physical quadrotor platform.

Abstract