Autonomous systems like unmanned aircraft systems are touted as front-runner in terms of efficiency and agility in challenging operations like emergency response. Ability of autonomous agents in such systems to avoid collision with dynamic obstacles while navigating through an
obstacle-cluttered environment is critical for success in any mission. This problem becomes more challenging in environments with uncertainties in obstacles motion, sensing and occluded regions.
To this end, a higher-order velocity-obstacle-based novel motion planner is presented in this thesis in a probabilistic set-up for smooth, collision-free navigation of the agent with acceleration
constraints in uncertain, unstructured environments with an element of anticipation for the future environment. Effectiveness of the developed algorithm for safe, collision-free, and smooth navigation of the agent is investigated in simulation studies in two different kinds of environments, one with known trajectories of obstacles and the other with unknown non-linear trajectories of obstacles. Extensive simulation studies are performed in the presence of dynamic obstacles to
elucidate the performance of the proposed algorithm on four crucial parameters - mission time, computational time, minimum obstacle distance, an overall control effort under varied obstacle
densities. With satisfactory performance in all these aspects, the developed algorithm possesses strong potential for real-time implementation.
However, it is important to note that the success rate of the proposed algorithm depends on the accuracy of prediction of future states of the obstacles by the Gaussian Processes (GP) algorithm. However, GP-based predictions are usually not interaction-aware and therefore, could not take into account the contextual cues in the environment like the presence of static physical obstacles, social conventions, etc. in its forecasting. Interaction-aware methods like Social Forces-based existing
methods consider all possible interactions. However, their effectiveness gets limited when interactions complexity increases. To this end, a statistical behavioural model, termed Social Gaussian Process (SGP) model, which combines the Social Force model (to capture the contextual cues and social conventions of movement in the presence of dynamic, static obstacles) with the Gaussian Processes (to learn the complex interactions using the past data reliably) in a Bayesian
framework is subsequently presented in this thesis. Predictions for the agents are further perfected by embedding the consideration of sensing and dynamic feasibility in the SGP formulation. Simulation results are presented to justify the betterment of performance of SGP than that of the GP-based one.