Obstacle avoidance and trajectory generation using artificial velocity function method

Abstract

There have been many methods that were researched in depth in Unamnned Aerial Vehicle trajectory generation for obstacle avoidance. These include solving for Laplace’s equations, wavefront propagation simulation method, distance functions and potential function methods. Howver, limited study have been done in the area of Artificial Velocity Method and its implementation in a off-the-shelf Hexacopter. This method uses a repulsive velocity field and attractive velocity field for the representation of obstacles and the goal point respectively. A physical space is modelled by the summation of the fields where the vectors at each point gives the speed and direction for the UAV to fly. The GPS dependency of the Hexacopter became a big hurdle for the algorithm to be succesfully implemented. Matlab was used to evaluate the performance of the method through velocity plots where shortfalls were identified. Improvements to the velocity equations were made with positive results.