Route Optimization via Environment-Aware Deep Network and Reinforcement Learning

Vehicle mobility optimization in urban areas is a long-standing problem smart city and spatial data analysis. Given the complex scenario unpredictable social events, our work focuses on developing mobile sequential recommendation system to maximize profitability of vehicle service providers (e.g., taxi drivers). In particular, we treat dynamic route as long-term decision-making task. A reinforcement-learning framework proposed tackle this problem, by integrating self-check mechanism deep neural network for customer pick-up point monitoring. To account unexpected situations COVID-19 outbreak), method designed be capable handling related environment changes with self-adaptive parameter determination mechanism. Based yellow New York City vicinity before after outbreak, have conducted comprehensive experiments evaluate effectiveness method. The results show consistently excellent performance, from hourly weekly measures, support superiority over state-of-the-art methods (i.e., more than 98% improvement terms