Dynamic Hybrid Path Planning using Grey Wolf Optimizer with Artificial Potential field for Autonomous UAVs
To make unmanned aerial vehicle (UAV) fully autonomous, path planning technology with sufficient performance is crucial. For a path planning algorithm to be significantly safer for a UAV, it needs to generate stable path with globally optimal cost and at shorter convergence iteration. Thus, this study proposes a hybrid path planning algorithm comprised of modified GWO using the novel approach of divide-and-conquer method for optimal global-path, which is called D&C GWO, given the known map with static objects, and improved dynamic artificial potential field, which is called A-APF to cope with the unknown moving objects which prompts UAV to avoid them in the context of global optimal path. Consequently, the significantly improved performance of our hybrid path planning algorithm was validated by conducting both simulation and real-world flight test.
Evaluation of Improved Maneuvering Performance using AGV-UAV Cooperative System in Virtual Environment Road Scenarios
The sight of a self-driving system in an autonomous ground vehicle (AGV), or the sensor system of an autonomous vehicle, is limited due to its low viewpoint. An autonomous unmanned aerial vehicle (UAV) flying within cooperative controlled with the AGV can provide much longer and wider bird-eye view map. By pairing an autonomous GV and a UAV, the system gains better access to blind spots rather than AGV itself. The path planning and tracking abilities of an AGV and UAV cooperative system are qualitatively evaluated upon multiple road scenarios in a virtual environment simulation. With the aid of UAV’s early detection of obstacles, the proposed system convinces that both the autonomous vehicle maneuvering driving performance and the possibility of the collision avoidance are substantially improved.
The following links would be the good references and videos produced by MathWorks Co.