This thesis tackles the challenge of integrating robots into human-robot collaboration in shared workspaces. It focuses on two key objectives: generating predictable robot motion during human interactions and designing a fast, optimal controller adaptable to environmental changes. The thesis begins by integrating human motion prediction into the obstacle avoidance behavior of the robot, enhancing its performance in confined spaces. It introduces a framework for generating predictable robot motion through interaction, further enhanced by a task generalization method. Lastly, TC-SAC controller, a fast and close-to-optimal controller for executing planned robot trajectories, is introduced.
«
This thesis tackles the challenge of integrating robots into human-robot collaboration in shared workspaces. It focuses on two key objectives: generating predictable robot motion during human interactions and designing a fast, optimal controller adaptable to environmental changes. The thesis begins by integrating human motion prediction into the obstacle avoidance behavior of the robot, enhancing its performance in confined spaces. It introduces a framework for generating predictable robot motio...
»