A Hybrid Navigation Framework for Autonomous Vehicles Based on A*-Stanley Integration and Real-Time UAV Perception

A Hybrid Navigation Framework for Autonomous Vehicles Based on A*-Stanley Integration and Real-Time UAV Perception

lida noorzadanzabi1

1) M.Sc. in Aerospace Engineering, Malek Ashtar University of Technology, Tehran, Iran

IOI Code : ACCA-ACAGBF
Publication : 4th International Congress on Science, engineering & New Technologies(secongress.com)
Abstract :
    In recent years, autonomous vehicles have been introduced as one of the most significant achievements in the field of intelligent transportation. However, they still face a serious challenge: limited field of view and the presence of blind spots in crowded urban environments. To address this perceptual gap, the present study proposes a hybrid navigation approach based on the use of Unmanned Aerial Vehicles (UAVs) as an aerial observer. In the designed framework, the UAV is responsible for detecting obstacles and transmitting their coordinates to the ground vehicle. Subsequently, a dual navigation system is implemented in MATLAB: the A* algorithm, by applying a 2-meter safety margin, generates an optimal and collision-free path, and the Stanley controller keeps the vehicle on this path by precisely adjusting the steering angle. To evaluate the performance, two different scenarios were simulated. In the first scenario, where the vehicle relied solely on its own sensors, collision with obstacles was inevitable; however, in the second scenario, where UAV data was available, the vehicle was able to traverse a completely safe path (with a 100% success rate). From a computational perspective, the A* algorithm generated the path in only 2.7 milliseconds, which attests to the system s capability for real-time execution. The findings of this research indicate that the integration of aerial perception with ground control is an effective step toward enhancing the safety of next-generation autonomous vehicles.

Keywords : autonomous vehicles uav assisted navigation a algorithm stanley controller collision avoidance matlab simulation