With a rapidly ageing population, innovative social care solutions are essential. This paper introduces an adaptive planning framework for assistive-care robotic missions, particularly focusing on mobile robots that interact with humans in home environments and provide support with everyday tasks. Our framework employs a modified version of Dijkstra’s algorithm and probabilistic model checking to ensure both efficient robot navigation for task support and human safety. Key contributions include the integration of formal verification techniques and adaptive path-finding methods, ensuring robust operation, even under the uncertainty that arises from runtime changes in the environment. Evaluation through simulation and real-world experiments with a mobile robotic platform PAL Robotics TIAGo demonstrates the effectiveness of the framework and the feasibility of our proposed approach, which contributes to the promotion of independent living for people with mild motor or cognitive impairments.
Stefanakos et al. (Tue,) studied this question.