Aiming at oscillation and stagnation problems generated by artificial potential field method

an improved method of feedforward control is proposed.Through analysis on the previous state

the expected value of the current state is calculated

and combined with the solution value of the current path planning algorithm

the final control amount of the current state is obtained

so that the traveling route of the drone is smoother

and the local optimal problem is also has been solved.In order to solve the collision problem between the clusters

the repulsion field function is layered and the machine is used as the obstacle point in real time.In the design of the group flight system

the host runs the algorithm

calculates the path points of the unmanned aerial vehicle（UAV）group in real time

and sends them to the corresponding slaves to complete the obstacle avoidance flight of the fleet.The results show that the scheme can meet the safety

real-time and stability requirements of the drone group.