A Hybrid Dynamic-Static Obstacle Avoidance Method for UAVs Based on Higher-Order Extended Control Obstacle Functions

Abstract

This invention relates to a hybrid obstacle avoidance method for UAVs based on a higher-order extended control obstacle function, belonging to the field of UAV obstacle avoidance technology. It includes the construction of a dynamic obstacle avoidance extended system, the design of a higher-order extended control obstacle function, the design of a nominal sliding mode controller, and the correction of the optimal controller using quadratic programming. By extending the obstacle motion state into a new system state through the design of a higher-order extended control obstacle function, complex nonlinear higher-order constraints are transformed into easily tractable linear constraints on the control input, significantly reducing the pressure of nonlinear constraints on real-time computation. Based on this, an obstacle avoidance constraint function is constructed for each obstacle, and these constraints are integrated into a unified optimization framework using the constraint integration capability of quadratic programming. This enables adaptive responses to environmental changes in scenarios where static and dynamic obstacles coexist, reducing obstacle avoidance response delay and trajectory oscillation risk, and improving the real-time performance of dynamic obstacle avoidance and its scalability in multi-obstacle environments.

Images