Model-Based Adaptation for Robotics Software

We developed model-based adaptation, an approach that leverages models of software and its environment to enable automated adaptation. The goal of our approach is to build long-lasting software systems that can effectively adapt to changes in their environment.

Mobile robotics is a growing area of research, but the long-term deployment of such robots has had limited success. A major challenge in the field is that robotics software is often purpose built for a given robot, task, and environment. Adapting that software to accommodate even small changes in the robot’s hardware, mission, or ecosystem is difficult. At Carnegie Mellon University (CMU), CoBots have been assisting humans for the last seven plus years by escorting visitors, delivering messages, and carrying out other tasks. But getting these robots to perform as intended requires a good deal of human effort. So the CoBots remain mostly confined to the controlled environment of the Computer Science building.