Context-based navigation for ground mobile robot in a semi-structured indoor environment
This work addresses the need for adaptable navigation in semi-structured indoor environments like retail, though it appears incremental as it builds on existing self-adaptation frameworks.
The paper tackles the problem of safe and efficient robot navigation in variable indoor environments by developing quality models that predict navigation safety based on the environment, enabling runtime adaptation of the local planner configuration. Experiments in a retail scenario validated the models and their integration into a self-adaptation framework.
There is a growing demand for mobile robots to operate in more variable environments, where guaranteeing safe robot navigation is a priority, in addition to time performance. To achieve this, current solutions for local planning use a specific configuration tuned to the characteristics of the application environment. In this paper, we present an approach for developing quality models that can be used by a self-adaptation framework to adapt the local planner configuration at run-time based on the perceived environment. We contribute a definition of a safety model that predicts the safety of a navigation configuration given the perceived environment. Experiments have been performed in a realistic navigation scenario for a retail application to validate the obtained models and demonstrate their integration in a self-adaptation framework.