Reinforcement Learning for Delivery Drone-Based Participatory Sensing in Dynamic Environments

Researchers propose a Two TimeScale Reinforcement Learning framework for delivery drone-based participatory sensing in dynamic environments. The framework, called TSRL, separates decision-making into macro and micro levels to tackle scalability and decision heterogeneity. TSRL achieves average system profit improvements of 20.1% in Hangzhou and 46.6% in Shanghai.