We explore the effects of robot failure severity (no failure vs. lowimpact vs. high-impact) on people's subjective ratings of the robot. We designed an escape room scenario in which one participant teams up with a remotely-controlled Pepper robot.We manipulated the robot's performance at the end of the game: The robot would either correctly follow the participant's instructions (control condition), the robot would fail but people could still complete the task of escaping the room (low-impact condition), or the robot's failure would cause the game to be lost (high-impact condition). Results showed no difference across conditions for people's ratings of the robot in terms of warmth, competence, and discomfort. However, people in the low-impact condition had significantly less faith in the robot's robustness in future escape room scenarios. Open-ended questions revealed interesting trends that are worth pursuing in the future: people may view task performance as a team effort and may blame their team or themselves more for the robot failure in case of a high-impact failure as compared to the low-impact failure.

A longstanding barrier to deploying robots in the real world is the ongoing need to author robot behavior. Remote data collection-particularly crowdsourcing-is increasingly receiving interest. In this paper, we make the argument to scale robot programming to the crowd and present an initial investigation of the feasibility of this proposed method. Using an off-the-shelf visual programming interface, non-experts created simple robot programs for two typical robot tasks (navigation and pick-and-place). Each needed four subtasks with an increasing number of programming statements (if statement, while loop, variables) for successful completion of the programs. Initial findings of an online study (N = 279) indicate that non-experts, after minimal instruction, were able to create simple programs using an off-the-shelf visual programming interface. We discuss our findings and identify future avenues for this line of research.

Reinforcement learning has shown great potential for learning sequential decision-making tasks. Yet, it is difficult to anticipate all possible real-world scenarios during training, causing robots to inevitably fail in the long run. Many of these failures are due to variations in the robot's environment. Usually experts are called to correct the robot's behavior; however, some of these failures do not necessarily require an expert to solve them. In this work, we query non-experts online for help and explore 1) if/how non-experts can provide feedback to the robot after a failure and 2) how the robot can use this feedback to avoid such failures in the future by generating shields that restrict or correct its high-level actions. We demonstrate our approach on common daily scenarios of a simulated kitchen robot. The results indicate that non-experts can indeed understand and repair robot failures. Our generated shields accelerate learning and improve data-efficiency during retraining.

Driving styles play a major role in the acceptance and use of autonomous vehicles. Yet, existing motion planning techniques can often only incorporate simple driving styles that are modeled by the developers of the planner and not tailored to the passenger. We present a new approach to encode human driving styles through the use of signal temporal logic and its robustness metrics. Specifically, we use a penalty structure that can be used in many motion planning frameworks, and calibrate its parameters to model different automated driving styles. We combine this penalty structure with a set of signal temporal logic formula, based on the Responsibility-Sensitive Safety model, to generate trajectories that we expected to correlate with three different driving styles: aggressive, neutral, and defensive. An online study showed that people perceived different parameterizations of the motion planner as unique driving styles, and that most people tend to prefer a more defensive automated driving style, which correlated to their self-reported driving style.

Data-driven robotic manipulation has been gaining traction. However, creating synthetic large-scale datasets for training, validation and benchmarks often relies on random sampling or perturbations, and the resulting scenarios do not necessarily reflect the desired task goals or spatial constraints on the manipulated objects, i.e., they are not spatially structured. We leverage spatial logics and generative models to automatically create spatially-structured manipulation scenarios from high-level specifications. We condition the models on such specifications to impose diverse spatial object relations on the data, e.g., the mug should be left of the plate. This approach enables users to define custom specifications and generate millions of scenarios within minutes, which specifically satisfy or violate the specifications to a desired extent.