Many real-world problems rely on the collaboration of multiple agents. Making plans for these multiple agents such that the goal state can be achieved becomes more and more difficult as the number of objects to consider increases. The increase in the number of objects results in the exponential increase in time and space required to find a viable plan. By mapping each agent onto some team, creating an abstract plan, and applying the abstract plan to the concrete problem, we can produce plans that reach the goal state more quickly than by solving them directly. This is demonstrated by applying this method to multiple problems in a custom domain dubbed the “garden” domain.

Understanding the limits and capabilities of an AI system is essential for safe and effective usability of modern AI systems. In the query-based AI assessment paradigm, a personalized assessment module queries a black-box AI system on behalf of a user and returns a user-interpretable model of the AI system’s capabilities. This thesis develops this paradigm to learn interpretable action models of simulator-based agents. Two types of agents are considered: the first uses high-level actions where the user’s vocabulary captures the simulator state perfectly, and the second operates on low-level actions where the user’s vocabulary captures only an abstraction of the simulator state. Methods are developed to interface the assessment module with these agents. Empirical results show that this method is capable of learning interpretable models of agents operating in a range of domains.

In a pursuit-evasion setup where one group of agents tracks down another adversarial group, vision-based algorithms have been known to make use of techniques such as Linear Dynamic Estimation to determine the probable future location of an evader in a given environment. This helps a pursuer attain an edge over the evader that has conventionally benefited from the uncertainty of the pursuit. The pursuer can utilize this knowledge to enable a faster capture of the evader, as opposed to a pursuer that only knows the evader's current location. Inspired by the function of dorsal anterior cingulate cortex (dACC) neurons in natural predators, the use of a predictive model that is built using an encoder-decoder Long Short-Term Memory (LSTM) Network and can produce a more accurate estimate of the evader's future location is proposed. This enables an even quicker capture of a target when compared to previously used filtering-based methods. The effectiveness of the approach is evaluated by setting up these agents in an environment based in the Modular Open Robots Simulation Engine (MORSE). Cross-domain adaptability of the method, without the explicit need to retrain the prediction model is demonstrated by evaluating it in another domain.

In settings where a human and an embodied AI (artificially intelligent) agent coexist, the AI agent has to be capable of reasoning with the human's preconceived notions about the environment as well as with the human's perception limitations. In addition, it should be capable of communicating intentions and objectives effectively to the human-in-the-loop. While acting in the presence of human observers, the AI agent can synthesize interpretable behaviors like explicable, legible, and assistive behaviors by accounting for the human's mental model (inclusive of her sensor model) in its reasoning process. This thesis will study different behavior synthesis algorithms which focus on improving the interpretability of the agent's behavior in the presence of a human observer. Further, this thesis will study how environment redesign strategies can be leveraged to improve the overall interpretability of the agent's behavior. At times, the agent's environment may also consist of purely adversarial entities or mixed entities (i.e. adversarial as well as cooperative entities), that are trying to infer information from the AI agent's behavior. In such settings, it is crucial for the agent to exhibit obfuscatory behavior that prevents sensitive information from falling into the hands of the adversarial entities. This thesis will show that it is possible to synthesize interpretable as well as obfuscatory behaviors using a single underlying algorithmic framework.

Most planning agents assume complete knowledge of the domain, which may not be the case in scenarios where certain domain knowledge is missing. This problem could be due to design flaws or arise from domain ramifications or qualifications. In such cases, planning algorithms could produce highly undesirable behaviors. Planning with incomplete domain knowledge is more challenging than partial observability in the sense that the planning agent is unaware of the existence of such knowledge, in contrast to it being just unobservable or partially observable. That is the difference between known unknowns and unknown unknowns.

In this thesis, I introduce and formulate this as the problem of Domain Concretization, which is inverse to domain abstraction studied extensively before. Furthermore, I present a solution that starts from the incomplete domain model provided to the agent by the designer and uses teacher traces from human users to determine the candidate model set under a minimalistic model assumption. A robust plan is then generated for the maximum probability of success under the set of candidate models. In addition to a standard search formulation in the model-space, I propose a sample-based search method and also an online version of it to improve search time. The solution presented has been evaluated on various International Planning Competition domains where incompleteness was introduced by deleting certain predicates from the complete domain model. The solution is also tested in a robot simulation domain to illustrate its effectiveness in handling incomplete domain knowledge. The results show that the plan generated by the algorithm increases the plan success rate without impacting action cost too much.

Many real-world planning problems can be modeled as Markov Decision Processes (MDPs) which provide a framework for handling uncertainty in outcomes of action executions. A solution to such a planning problem is a policy that handles possible contingencies that could arise during execution. MDP solvers typically construct policies for a problem instance without re-using information from previously solved instances. Research in generalized planning has demonstrated the utility of constructing algorithm-like plans that reuse such information. However, using such techniques in an MDP setting has not been adequately explored.

This thesis presents a novel approach for learning generalized partial policies that can be used to solve problems with different object names and/or object quantities using very few example policies for learning. This approach uses abstraction for state representation, which allows the identification of patterns in solutions such as loops that are agnostic to problem-specific properties. This thesis also presents some theoretical results related to the uniqueness and succinctness of the policies computed using such a representation. The presented algorithm can be used as fast, yet greedy and incomplete method for policy computation while falling back to a complete policy search algorithm when needed. Extensive empirical evaluation on discrete MDP benchmarks shows that this approach generalizes effectively and is often able to solve problems much faster than existing state-of-art discrete MDP solvers. Finally, the practical applicability of this approach is demonstrated by incorporating it in an anytime stochastic task and motion planning framework to successfully construct free-standing tower structures using Keva planks.

Robot motion planning requires computing a sequence of waypoints from an initial configuration of the robot to the goal configuration. Solving a motion planning problem optimally is proven to be NP-Complete. Sampling-based motion planners efficiently compute an approximation of the optimal solution. They sample the configuration space uniformly and hence fail to sample regions of the environment that have narrow passages or pinch points. These critical regions are analogous to landmarks from planning literature as the robot is required to pass through them to reach the goal.

This work proposes a deep learning approach that identifies critical regions in the environment and learns a sampling distribution to effectively sample them in high dimensional configuration spaces.

A classification-based approach is used to learn the distributions. The robot degrees of freedom (DOF) limits are binned and a distribution is generated from sampling motion plan solutions. Conditional information like goal configuration and robot location encoded in the network inputs showcase the network learning to bias the identified critical regions towards the goal configuration. Empirical evaluations are performed against the state of the art sampling-based motion planners on a variety of tasks requiring the robot to pass through critical regions. An empirical analysis of robotic systems with three to eight degrees of freedom indicates that this approach effectively improves planning performance.

In order for a robot to solve complex tasks in real world, it needs to compute discrete, high-level strategies that can be translated into continuous movement trajectories. These problems become increasingly difficult with increasing numbers of objects and domain constraints, as well as with the increasing degrees of freedom of robotic manipulator arms.

The first part of this thesis develops and investigates new methods for addressing these problems through hierarchical task and motion planning for manipulation with a focus on autonomous construction of free-standing structures using precision-cut planks. These planks can be arranged in various orientations to design complex structures; reliably and autonomously building such structures from scratch is computationally intractable due to the long planning horizon and the infinite branching factor of possible grasps and placements that the robot could make.

An abstract representation is developed for this class of problems and show how pose generators can be used to autonomously compute feasible robot motion plans for constructing a given structure. The approach was evaluated through simulation and on a real ABB YuMi robot. Results show that hierarchical algorithms for planning can effectively overcome the computational barriers to solving such problems.

The second part of this thesis proposes a deep learning-based algorithm to identify critical regions for motion planning. Further investigation is done whether these learned critical regions can be translated to learn high-level landmark actions for automated planning.

This research introduces Roblocks, a user-friendly system for learning Artificial Intelligence (AI) planning concepts using mobile manipulator robots. It uses a visual programming interface based on block-structured programming to make AI planning concepts easier to grasp for those who are new to robotics and AI planning. Users get to accomplish any desired tasks by dynamically populating puzzle shaped blocks encoding the robot’s possible actions, allowing them to carry out tasks like navigation, planning, and manipulation by connecting blocks instead of writing code. Roblocks has two levels, where in the first level users are made to re-arrange a jumbled set of actions of a plan in the correct order so that a given goal could be achieved. In the second level, they select actions of their choice but at each step only those actions pertaining to the current state are made available to them, thereby pruning down the vast number of possible actions and suggesting only the truly feasible and relevant actions. Both of these levels have a simulation where the user plan is executed. Moreover, if the user plan is invalid or fails to achieve the given goal condition then an explanation for the failure is provided in simple English language. This makes it easier for everyone (especially for non-roboticists) to understand the cause of the failure.

Facial Expressions Recognition using the Convolution Neural Network has been actively researched upon in the last decade due to its high number of applications in the human-computer interaction domain. As Convolution Neural Networks have the exceptional ability to learn, they outperform the methods using handcrafted features. Though the state-of-the-art models achieve high accuracy on the lab-controlled images, they still struggle for the wild expressions. Wild expressions are captured in a real-world setting and have natural expressions. Wild databases have many challenges such as occlusion, variations in lighting conditions and head poses. In this work, I address these challenges and propose a new model containing a Hybrid Convolutional Neural Network with a Fusion Layer. The Fusion Layer utilizes a combination of the knowledge obtained from two different domains for enhanced feature extraction from the in-the-wild images. I tested my network on two publicly available in-the-wild datasets namely RAF-DB and AffectNet. Next, I tested my trained model on CK+ dataset for the cross-database evaluation study. I prove that my model achieves comparable results with state-of-the-art methods. I argue that it can perform well on such datasets because it learns the features from two different domains rather than a single domain. Last, I present a real-time facial expression recognition system as a part of this work where the images are captured in real-time using laptop camera and passed to the model for obtaining a facial expression label for it. It indicates that the proposed model has low processing time and can produce output almost instantly.