Insights from autonomy trials of a self-reconfigurable floor-cleaning robot in a public food court

As autonomous cleaning robots advance, we expect certain features, such as higher area coverage and robustness. To explore these features and their challenges, we need tools and strategies that allow us to develop them rapidly. In this paper, we present lessons learned and results while performing 3 months long field trials on the autonomy of a self-reconfigurable floor-cleaning robot. The autonomy here includes path planning and navigation of the robot for coverage in a public food court. The environment of the food court is often subjected to alterations due to human activities. One of the main contributions of this paper is proposing a task-level execution system to deploy the cleaning robot in an actual use case scenario. We also provide in-depth discussions on the solutions and the implementation strategies we found in accomplishing goals. These findings would be helpful in conducting autonomy development targeted for field deployments of self-reconfigurable robots and cleaning robots.