While deep reinforcement learning (RL) has been increasingly applied in designing car-following models in the last years, this study aims at investigating the feasibility of RL-based vehicle-following for complex vehicle dynamics and strong environmental disturbances. As a use case, we developed an inland waterways vessel-following model based on realistic vessel dynamics, which considers environmental influences, such as varying stream velocity and river profile. We extracted natural vessel behavior from anonymized AIS data to formulate a reward function that reflects a realistic driving style next to comfortable and safe navigation. Aiming at high generalization capabilities, we propose an RL training environment that uses stochastic processes to model leading trajectory and river dynamics. To validate the trained model, we defined different scenarios that have not been seen in training, including realistic vessel-following on the Middle Rhine. Our model demonstrated safe and comfortable driving in all scenarios, proving excellent generalization abilities. Furthermore, traffic oscillations could effectively be dampened by deploying the trained model on a sequence of following vessels.
Vessel-following model for inland waterways based on deep reinforcement learning
2022
Preprint
Electronic Resource
English
Optimal Vessel Planning in Natural Inland Waterways
Springer Verlag | 2024
|Engineering Index Backfile | 1924
|Engineering Index Backfile | 1929
|INLAND WATERWAYS The Inland Waterways capital development plan
Online Contents | 2010
Safe Optimal Vessel Planning on Natural Inland Waterways
IEEE | 2023
|