Leveraging Deep Reinforcement Learning and Healthcare Devices for Active Travelling in Smart Cities

Smart cities are increasingly challenged by population growth and the environmental emissions of urban transportation systems, necessitating sustainable urban planning to improve public health, environmental quality, and overall urban livability. A notable aspect in this context is the under-utilization of smart healthcare wearable devices or smart healthcare applications in urban transportation systems. This paper proposes an innovative approach to address these challenges effectively. We formulate a non-convex optimization problem aimed at minimizing environmental emissions within transportation systems while considering resident health goals, travel time constraints, and infrastructure limitations. To achieve this, we employ deep reinforcement learning (DRL), which dynamically selects the optimal traveling mode for residents. This approach aims to optimize environmental outcomes while meeting individualized mobility needs. Moreover, our method integrates smart healthcare technologies to capture real-time data and predict optimal traveling modes. By incorporating real-world health metrics into transportation planning, we enhance decision-making processes and promote active transportation options, contributing to healthier urban environments. Through extensive simulations, we demonstrate the effectiveness of our approach in optimizing traveling decisions and advancing sustainable urban mobility practices. Our DRL-based solution effectively promotes active travel, leading to a significant increase in health-related metrics (like calories burned) and a substantial reduction in gCO2 emissions. Up to 74% of journeys were made using active transportation modes. Cycling is particularly popular, accounting for up to 67% of journeys.