Deep-Dispatch: A Deep Reinforcement Learning-Based Vehicle Dispatch Algorithm for Advanced Air Mobility
This addresses the practical challenge of efficient vehicle dispatch for advanced air mobility operators, offering a scalable solution to overcome computational bottlenecks in optimization models.
The paper tackles the eVTOL dispatch problem for air taxi operations, which is constrained by recharging needs, limited vertiports, and dynamic demand and electricity prices, by developing deep reinforcement learning algorithms that maximize operating profit. The multi-agent algorithm approximates the optimal policy with significantly less computational expense compared to benchmark optimization models, outperforming the single-agent version in profits and training time across 36 numerical cases.
Near future air taxi operations with electric vertical take-off and landing (eVTOL) aircraft will be constrained by the need for frequent recharging of eVTOLs, limited takeoff and landing pads in vertiports, and subject to time-varying demand and electricity prices, making the eVTOL dispatch problem unique and particularly challenging to solve. Previously, we have developed optimization models to address this problem. Such optimization models however suffer from prohibitively high computational run times when the scale of the problem increases, making them less practical for real world implementation. To overcome this issue, we have developed two deep reinforcement learning-based eVTOL dispatch algorithms, namely single-agent and multi-agent deep Q-learning eVTOL dispatch algorithms, where the objective is to maximize operating profit. An eVTOL-based passenger transportation simulation environment was built to assess the performance of our algorithms across $36$ numerical cases with varying number of eVTOLs, vertiports, and demand. The results indicate that the multi-agent eVTOL dispatch algorithm can closely approximate the optimal dispatch policy with significantly less computational expenses compared to the benchmark optimization model. The multi-agent algorithm was found to outperform the single-agent counterpart with respect to both profits generated and training time.