University Archives

As part of the Electrical and Computer Engineering Senior Design course, the Navigation and Controls team is developing the subsystem responsible for enabling Level 5 autonomy for a ground vehicle competing in the 2026 Intelligent Ground Vehicle Competition (IGVC) AutoNav Challenge. The autonomous platform must navigate to predefined GPS waypoints while detecting and avoiding obstacles in an unknown environment. To accomplish this, a custom path planning framework was developed within the Robotic Operating System 2 (ROS2) environment.

The planner leverages an Artificial Potential Field (APF) approach with a costmap-based representation of the environment generated from LiDAR and localization data. The APF method guides the vehicle toward the goal while repulsive potentials prevent collisions with obstacles. Candidate motion options are generated using a circular sampling method, allowing the planner to evaluate multiple possible directions of motion and improving robustness against local minima. When the planner encounters a navigational trap, a simulated annealing–based escape mechanism is used to enable the system to escape local minima and continue progressing toward the goal.

The combination of APF guidance, circular sampling, and simulated annealing enables reliable obstacle avoidance and goal-directed navigation in complex environments. Current work focuses on system integration and validation prior to field testing.