Path-Planning and Optimization of a High-speed 1/5-Scale Off-Road Autonomous Vehicle

Open Access
- Author:
- Esparragoza, Andres
- Area of Honors:
- Mechanical Engineering
- Degree:
- Bachelor of Science
- Document Type:
- Thesis
- Thesis Supervisors:
- Sean N Brennan, Thesis Supervisor
Daniel Humberto Cortes Correales, Thesis Honors Advisor - Keywords:
- Autonomous Vehicles
Off-Road Vehicles
High-Speed
Teensy 4.1
Driving Simulation
Iterative Learning Control
Steady-State Velocity Profile
Encoders
GPS - Abstract:
- This thesis investigates path-planning for a high-speed off-road autonomous vehicle. A 1/5th scale Remote Control (RC) vehicle was modified with sensors to be able to track vehicle position, trajectory, velocity, and other parameters necessary while performing traversals on a specifically designed off-road course that challenges the vehicles maneuverability and velocity. Advanced global position system (GPS) and encoder algorithms are implemented into the vehicles onboard microcontrollers to determine real time data necessary to dictate the necessary steering and throttle inputs. The sensor algorithms allow for the vehicle to measure its position and velocity, thereby enabling the foundation of path-following. Using algorithms to control the two receiving functions, steering and throttle, the vehicle must traverse the course autonomously with the aim of increasing speed per traversal until reaching an optimal steady-state velocity profile. Using iterative learning control (ILC) and similar control strategies, the vehicle learns from prior traversals to find limiting levels of velocity to be able to repeatedly produce traversals faster with each lap. The foundation for a 1/5th scale high-speed off-road vehicle path following simulation is created on MATLAB prior to vehicle implementation. The simulation successfully models constant speed along a path for both large-scale and 1/5th scale vehicles. The simulation is prepared for the implementation of ILC to determine the optimal steady-state velocity profile. The goal is for the simulation to achieve lap navigation faster than if a human were to control the vehicle via its remote-control unit. This research discussion concludes by outlining the next steps for algorithm development and field testing. The overall goal is to contribute to the development of autonomous off-road vehicles for use in a range of applications, including agriculture, mining, and search and rescue. The findings of this thesis seek to advance core algorithms and deployment technologies for autonomous systems in challenging off-road environments.