A New Multidimensional Repulsive Potential Field to Avoid Obstacles by Nonholonomic UAVs in Dynamic Environments.

Area of Science:

• Robotics
• Aerospace Engineering
• Control Systems

Background:

• Autonomous flight with obstacle avoidance is crucial for modern unmanned aerial vehicles (UAVs).
• Nonholonomic UAVs, especially fixed-wing aircraft, face increased complexity in avoiding moving obstacles due to limited maneuverability.
• Traditional repulsive potential fields are inadequate for nonholonomic vehicles as they do not account for obstacle velocity and UAV dynamics.

Purpose of the Study:

• To develop a novel multidimensional repulsive potential field specifically designed for nonholonomic UAVs.
• To enhance obstacle avoidance capabilities for UAVs operating in environments with moving obstacles.
• To address the limitations of traditional potential field methods in handling nonholonomic constraints and dynamic environments.

Main Methods:

• Introduction of a multidimensional repulsive potential field that generates forces based on the obstacle's velocity vector.
• The repulsive force is directed away from areas in front of and behind the obstacle along its movement path.
• Force magnitude is determined by the UAV's proximity to the obstacle's movement line, distance along the line, and relative projected speed.

Main Results:

• The proposed potential field successfully repulses the UAV from critical areas relative to the obstacle's motion.
• The multidimensional nature of the field accounts for relative position, speed, and direction of both UAV and obstacle.
• Numerical simulations validate the effectiveness of the novel potential field in controlling nonholonomic UAV flight.

Conclusions:

• The developed multidimensional repulsive potential field offers a significant improvement for obstacle avoidance in nonholonomic UAVs.
• This method provides a more robust and adaptable solution compared to traditional potential field approaches.
• The findings contribute to safer and more efficient autonomous navigation for UAVs in complex, dynamic scenarios.