Modern robots make decisions in many ways, but rely on their designers to choose which strategies to employ and when. Adopting a perspective of bounded rationality from the cognitive sciences, we develop a definition of decision making for constructivist robots to formulate their own decisions based on their mission-specific values. Our model, extending Kirsch (2019), defines decision making with an iterative algorithm that captures a broad range of possible strategies and problem domains. Briefly, a set of decision alternatives are first assessed by a set of relevant cues. These assessments are then aggregated into a multi-dimensional evaluation of each alternative from which a preference ordering is created. Finally, based on the problem specification, a set of chosen alternatives is either accepted by a stopping rule or a new iteration is started, updating the sets of alternatives and cues. If no alternatives or cues remain after iterating, then the decision fails. Given this algorithm, we implement a toolbox of decision-making components as modules in a cognitive robot architecture and demonstrate a method of assembling them into complete decision strategies, represented as behavior trees, using an automated theorem prover. For proof of concept, we simulate three example strategies used by a domestic robot performing a search and retrieval task. We discuss new insights into designing and selecting decision-making strategies and make recommendations on how our proof of concept can be improved.

In household and retail store environments, humans efficiently identify grasp regions of an object to perform everyday tasks. To enable robots to understand the requirements of a task and the properties of an object, a novel grasp framework, called the Shape Primitive and Reasoning Grasping Engine (SPaRGE), is proposed in this thesis. SPaRGE combines a data-driven approach, a superquadric algorithm, and a logic-based approach to find task-specific object grasp poses from a partial point cloud. Multiple grasp poses are recovered using the data-driven approach, while objects are represented as multiple primitive shapes using the superquadric algorithm and logic-based approach. Logic rules are used to decide which primitive shape meets the criteria of a given task, resulting in a primitive shape indicating a grasp region. The data-driven approach then finds a grasp pose that meets the task requirements. Two versions of the logic-based approach are presented, differing in their design choices that affect the generalizability and scalability of the logic rules. The SPaRGE framework is evaluated on common household and retail store objects, demonstrating an average true positive rate of 81.2% and 86.0%, respectively, in enabling a robot to grasp objects at a desired region. Results show that the SPaRGE framework enables real-time object grasping without prior knowledge of the objects. Further experiments show that the SPaRGE framework can be adapted successfully for real-world object grasping scenarios. The proposed framework provides a significant contribution toward enabling robots to perform human-like grasping in real-world environments.