Method for optimized autonomous drone swarm path planning
Methods and systems for unmaned aerial vehicles (UAV) swarm path planning based on potential field optimization. The method incorporates considerations of inter-UAV interaction forces, a collision risk calculation mechanism based on safe distance angles, and an auxiliary target point strategy. The system calculates collision risk by linearly dividing the safe distance angle between the UAV and obstacles. When the risk exceeds a threshold, the system generates two auxiliary sub-target points and selects the point closest to the flight start point as the optimal sub-target, guiding the UAV to avoid obstacles by applying auxiliary gravity. When the risk is below the threshold, the UAV is unaffected by repulsive forces. Therefore, frequency of heading angle changes of the drone in a swarm drone group is reduced based on the collision risk value comparisons.
1 . A method for a drone swarm path planning, wherein the method is executed by a vehicle with one or more processors, the method comprising:
acquiring state information of the drone and its environmental obstacles;
calculating a respective collision risk value based on a safe distance angle between the drone and each of the environmental obstacles;
comparing each of the respective collision risk values with a preset safety threshold;
in response to the each of the collision risk value exceeding the preset safety threshold:
determining an optimal auxiliary sub-target point; and applying an auxiliary gravitational force to the drone pointing towards the optimal auxiliary sub-target point to guide the drone to avoid obstacles along a tangential path;
in response to the each of the collision risk value being below the preset safety threshold:
configuring the drone to ignore a repulsive force effect from the environmental obstacles;
wherein frequency of heading angle changes of the drone is reduced based on the collision risk value comparisons;
wherein the environmental obstacle comprise one or neighboring drones in a drone swarm group in which the drone belongs to;
wherein the calculating the collision risk value further comprising normalizing the safe distance angle based on a preset linear division rule, such that the collision risk value varies linearly within the range of 0 to 1;
wherein the determining the optimal auxiliary sub-target point further comprises:
constructing a perpendicular line based on a connection between a current position of the drone and a position of an environmental obstacle;
determining two intersection points of the perpendicular line with a potential field boundary of the environmental obstacle as candidate auxiliary sub-target points;
calculating distances from two candidate auxiliary sub-target points to a flight start point of the drone; and
selecting the candidate auxiliary sub-target point closest to the flight start point as the optimal auxiliary sub-target point.