Goal-Oriented End-User Programming of Robots

End-user programming (EUP) tools must balance user control with the robot'sability to plan and act autonomously. Many existing task-oriented EUP toolsenforce a specific level of control, e.g., by requiring that users hand-craftdetailed sequences of actions, rather than offering users the flexibility tochoose the level of task detail they wish to express. We thereby created anovel EUP system, Polaris, that in contrast to most existing EUP tools, usesgoal predicates as the fundamental building block of programs. Users canthereby express high-level robot objectives or lower-level checkpoints at theirchoosing, while an off-the-shelf task planner fills in any remaining programdetail. To ensure that goal-specified programs adhere to user expectations ofrobot behavior, Polaris is equipped with a Plan Visualizer that exposes theplanner's output to the user before runtime. In what follows, we describe ourdesign of Polaris and its evaluation with 32 human participants. Our resultssupport the Plan Visualizer's ability to help users craft higher-qualityprograms. Furthermore, there are strong associations between user perception ofthe robot and Plan Visualizer usage, and evidence that robot familiarity has akey role in shaping user experience.