Deep Reinforcement Learning for Trajectory Control of Connected and Automated Vehicles at a Mixed-Traffic Intersection
Bibliographic record
Abstract
Vehicle trajectory control has garnered significant interest due to the potential of connected and automated vehicles (CAVs) to enhance traffic efficiency and reduce accidents. Effective vehicle control is crucial for autonomous driving and has traditionally been addressed through motion control methods. In recent years, a growing focus has been on deep learning (DL), particularly deep reinforcement learning (DRL), to optimize CAV control. These advanced techniques enable CAVs to learn and adapt to their environment, further advancing autonomous driving capabilities. However, much research primarily focuses on simple and ideal scenarios involving fully CAV environments. Although there is research on mixed-traffic scenarios with both CAVs and human-driven vehicles (HDVs), it often neglects other road participants such as pedestrians and cyclists. Many studies on trajectory control focus on enhancing traffic efficiency and reducing vehicle emissions. However, it is equally essential to consider safety improvements in mixed-traffic environments. Addressing the aforementioned issues, this paper proposes a DRL-based trajectory control approach for CAVs at a mixed-traffic intersection involving CAVs, HDVs, and pedestrians. CAVs learn policies for various actions to reach their destinations. The acceleration of the CAVs is optimized using the deep deterministic policy gradient (DDPG) algorithm to maximize a reward function that accounts for safety, energy efficiency, and traffic efficiency. Our approach is tested using the Simulation of Urban Mobility (SUMO) platform to model realistic intersection dynamics. Results demonstrated that the proposed method significantly improved traffic efficiency and reduced fuel consumption, while notably decreasing both vehicle–vehicle and vehicle–pedestrian conflicts. This research highlights the importance of incorporating diverse road users into CAV control strategies and contributes to the development of safer, greener, and more inclusive traffic management solutions.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".