Date of Award
Program or Major
Doctor of Philosophy
The scope of this work spans two main areas of autonomy research 1) autonomous path planning and 2) test and evaluation of autonomous systems. Path planning is an integral part of autonomous decision-making, and a deep understanding in this area provides valuable perspective on approaching the problem of how to effectively evaluate vehicle behavior.
Autonomous decision-making capabilities must include reliability, robustness, and trustworthiness in a real-world environment. A major component of robot decision-making lies in intelligent path-planning. Serving as the brains of an autonomous system, an efficient and reliable path planner is crucial to mission success and overall safety. A hybrid global and local planner is implemented using a combination of the Potential Field Method (PFM) and A-star (A*) algorithms. Created using a layered vector field strategy, this allows for flexibility along with the ability to add and remove layers to take into account other parameters such as currents, wind, dynamics, and the International Regulations for Preventing Collisions at Sea (COLGREGS). Different weights can be attributed to each layer based on the determined level of importance in a hierarchical manner. Different obstacle scenarios are shown in simulation, and proof-of-concept validation of the path-planning algorithms on an actual ASV is accomplished in an indoor environment. Results show that the combination of PFM and A* complement each other to generate a successfully planned path to goal that alleviates local minima and entrapment issues. Additionally, the planner demonstrates the ability to update for new obstacles in real time using an obstacle detection sensor.
Regarding test and evaluation of autonomous vehicles, trust and confidence in autonomous behavior is required to send autonomous vehicles into operational missions. The author introduces the Performance Evaluation and Review Framework Of Robotic Missions (PERFORM), a framework for which to enable a rigorous and replicable autonomy test environment, thereby filling the void between that of merely simulating autonomy and that of completing true field missions. A generic architecture for defining the missions under test is proposed and a unique Interval Type-2 Fuzzy Logic approach is used as the foundation for the mathematically rigorous autonomy evaluation framework. The test environment is designed to aid in (1) new technology development (i.e. providing direct comparisons and quantitative evaluations of varying autonomy algorithms), (2) the validation of the performance of specific autonomous platforms, and (3) the selection of the appropriate robotic platform(s) for a given mission type (e.g. for surveying, surveillance, search and rescue). Several case studies are presented to apply the metric to various test scenarios. Results demonstrate the flexibility of the technique with the ability to tailor tests to the user’s design requirements accounting for different priorities related to acceptable risks and goals of a given mission.
Dalpe, Allisa, "PERFORMANCE EVALUATION AND REVIEW FRAMEWORK OF ROBOTIC MISSIONS (PERFORM): AUTONOMOUS PATH PLANNING AND AUTONOMY PERFORMANCE EVALUATION" (2021). Doctoral Dissertations. 2563.