A Complete Autonomous Vehicle Solution: An ASV Case Study

The field of robotics has grown at an astonishing rate over the past two decades, and the interest in autonomous vehicles has grown even faster. Autonomous vehicles offer the unique promise of removing drivers from dangerous environments, and increasing the efficiency of our highways. However, the development of these systems is no trivial task. For an autonomous vehicle to effectively navigate and interact with the world, the vehicle must ( 1) determine where it is located and create an accurate assessment of its surroundings, (2) develop a feasible navigational path based on the current surroundings and the vehicle's destination, and (3) monitor the vehicle's progress as it moves through the environment. A deficiency in any of these requirements leads to results that are less desirable, and often dangerous. This research presents a complete solution for designing and building an autonomous vehicle. It describes the overall concepts inherent in any autonomous system, and then illustrates how the theoretical concepts can be utilized in a real world application. This work was conducted parallel to the development of Old Dominion University's Autonomous Surface Vehicle, Lil' Blue. The research reviews the kinematic equations of motion needed to model the physical system, and introduces the Kalman filter which is used to assess the vehicle's state information. A vision processing algorithm lets the boat assess its surroundings, a navigational path planning algorithm determines the boat's target destination, and a unique PID controller eliminates the need for complex user commands to navigate towards the desired target point. The algorithms where physically tested throughout the ASV's development, and those results are presented in this work. The overall goal of this research is to outline the necessary components to develop an autonomous vehicle, and then demonstrate those precepts with a real world application. The result of this research is a fully functional ASV, which is well equipped to compete in the next AUVSI RoboBoat competition.