This thesis discusses two different problems in motion planning for autonomous driving. The first is the problem of optimizing a lattice planner control set for any particular autonomous driving task, with the goal of reducing planning time for that task. The driving task is encoded in the form of a dataset of trajectories executed while performing said task. In addition to improving planning time, the optimized control set should capture the driving style of the dataset. In this sense, the control set is learned from the data and is tailored to a particular task. To determine the value of control actions to add to the control set, a modified version of the Fréchet distance is used to score how useful control actions are for generating paths similar to those in the dataset. This method is then compared to the state of the art lattice planner control set optimization technique in terms of planning runtime for the learned task. The second problem is the task of extending the Responsibility-Sensitive Safety (RSS) framework by introducing swerve manoeuvres in addition to the nominal braking manoeu- vres present in the framework. This includes comparing the clearance distances required by a swerve to the braking distances in the original framework. This comparison shows that swerve manoeuvres require less distance gap in order to reach safety from a braking agent in front of the autonomous vehicle at higher speeds. For more realistic swerve manoeuvres, the kinematic bicycle model is used rather than the 2-D double integrator model consid- ered in RSS. An upper bound is then computed on the required clearance distance for a swerve manoeuvre that satisfies bicycle kinematics. A longitudinal safe following distance is then derived that is provably safe, and is shown to be lower than the following distance required by RSS at higher speeds. The use of the kinematic bicycle model is then validated by computing swerve manoeuvres with a dynamic single-track car model and Pacejka tire model, and comparing the single-track swerves to the bicycle swerves.

Path Planning (PP) is one of the prerequisites in ensuring safe navigation and manoeuvrability control for driverless vehicles. Due to the dynamic nature of the real world, PP needs to address changing environments and how autonomous vehicles respond to them. This book explores PP in the context of road vehicles, robots, off-road scenarios, multi-robot motion, and unmanned aerial vehicles (UAVs ).

In motion planning for automated vehicles, a thorough uncertainty consideration is crucial to facilitate safe and convenient driving behavior. This work presents three motion planning approaches which are targeted towards the predominant uncertainties in different scenarios, along with an extended safety verification framework. The approaches consider uncertainties from imperfect perception, occlusions and limited sensor range, and also those in the behavior of other traffic participants.

This work develops a motion planner that compensates the deficiencies from perception modules by exploiting the reaction capabilities of a vehicle. The work analyzes present uncertainties and defines driving objectives together with constraints that ensure safety. The resulting problem is solved in real-time, in two distinct ways: first, with nonlinear optimization, and secondly, by framing it as a partially observable Markov decision process and approximating the solution with sampling.

The major problem addressed by this research is how to plan a safe motion for autonomous vehicles in a two dimensional, rectilinear world. With given start and goal configurations, the planner performs motion planning which will lead a vehicle to achieve its task safely. During the planning, in addition to the safety consideration, motion's smoothness is also taken into account. The approach taken was to divide whole motion planning task into two layers. The top layer finds a global path by decomposing the free space into convex regions, then searching for an optimal global path class. The bottom layer performs local motion planning which further subdivides the planning problem into mid-portion and end-portion motion planning. The local motion planning is carried out region by region along the global path class. As results, simple motion instructions are generated for each region. For execution of planned motion, a software system, Model-based Mobile robot Language (MML- 11), was developed. This easy- to-use robot language provides users a convenient tool to program their applications through its function library. The results of the research were implemented in MML-1l and tested on an experimental robot Yamabico-11 successfully. (AN).

In the past decade, the autonomous driving industry has seen tremendous advancements thanks to the progress in computation, artificial intelligence, sensing capabilities, and other technologies related to autonomous vehicles. Today, autonomous cars operate in dense urban traffic, compared to the last generation of robots that were confined to isolated workspaces. In these human-populated environments, autonomous cars need to understand their surroundings and behave in an interpretable, human-like manner. In addition, autonomous robots are engaged in more social interactions with other humans, which requires an understanding of how multiple reactive agents act. For example, during lane changes, most attentive drivers would slow down to give space if an adjacent car shows signs of executing a lane change. For an autonomous car, understanding the mutual dependence between its action and others' actions is essential for the safety and viability of the autonomous driving industry. However, most existing trajectory planning approaches ignore the coupling between all agents' behaviors and treat the decisions of other agents as immutable. As a result, the planned trajectories are conservative, less intuitive, and may lead to unsafe behaviors. To address these challenges, we present motion planning frameworks that maintain the coupling of prediction and planning by explicitly modeling their mutual dependency. In the first part, we examine reciprocal collision avoidance behaviors among a group of intelligent robots. We propose a distributed, real-time collision avoidance algorithm based on Voronoi diagrams that only requires relative position measurements from onboard sensors. When necessary, the proposed controller minimally modifies a nominal control input and provides collision avoidance behaviors even with noisy sensor measurements. In the second part, we introduce a nonlinear receding horizon game-theoretic planner that approximates a Nash equilibrium in competitive scenarios among multiple cars. The proposed planner uses a sensitivity-enhanced objective function and iteratively plans for the ego vehicle and the other vehicles to reach an equilibrium strategy. The resulting trajectories show that the ego vehicle can leverage its influence on other vehicles' decisions and intentionally change their courses. The resulting trajectories exhibit rich interactive behaviors, such as blocking and overtaking in competitive scenarios among multiple cars. In the last part, we propose a risk-aware game-theoretic planner that takes into account uncertainties of the future trajectories. We propose an iterative dynamic programming algorithm to solve a feedback equilibrium strategy set for interacting agents with different risk sensitivities. Through simulations, we show that risk-aware planners generate safer behaviors when facing uncertainties in safety-critical situations. We also present a solution for the "inverse" risk-sensitive planning algorithm. The goal of the inverse problem is to learn the cost function as well as risk sensitivity for each individual. The proposed algorithm learns the cost function parameters from datasets collected from demonstrations with various risk sensitivity. Using the learned cost function, the ego vehicle can estimate the risk profile of an interacting agent online to improve safety and efficiency.

The major problem addressed by this research is how to plan a safe motion for autonomous vehicles in a two dimensional, rectilinear world. With given start and goal configurations, the planner performs motion planning which will lead a vehicle to achieve its task safely. During the planning, in addition to the safety consideration, motion's smoothness is also taken into account. The approach taken was to divide whole motion planning task into two layers. The top layer finds a global path by decomposing the free space into convex regions, then searching for an optimal global path class. The bottom layer performs local motion planning which further subdivides the planning problem into mid-portion and end-portion motion planning. The local motion planning is carried out region by region along the global path class. As results, simple motion instructions are generated for each region. For execution of planned motion, a software system, Model-based Mobile robot Language (MML- 11), was developed. This easy- to-use robot language provides users a convenient tool to program their applications through its function library. The results of the research were implemented in MML-1l and tested on an experimental robot Yamabico-11 successfully. (AN).