Lane merging in autonomous vehicle urban driving using reinforcement learning models
Bibliographic record
Abstract
Autonomous vehicle lane merging is a critical task in urban driving, requiring precise navigation through complex and dynamic traffic environments. Challenges such as roadworks, lane reductions, merging from gas stations, low-visibility conditions, and crowded highway on-ramps demand continuous improvements in autonomous driving systems. Effective navigation in these situations, particularly at multi-lane junctions, merging onto high-speed roads, avoiding obstacles, and managing emergency vehicle lanes, requires robust decision-making that can adapt to changing road conditions. This paper compares three popular reinforcement learning (RL) algorithms—Proximal Policy Optimization (PPO), Advantage Actor-Critic (A2C), and Deep Q-Learning (DQL)—to address these challenges. Our findings show that in environments with specific decision points, DQL excels in tasks like lane reduction and obstacle avoidance due to its value-based approach. The A2C model, as an actor-critic policy, is particularly effective in dynamic environments, enabling the optimization of urban traffic control and merging at roundabouts. PPO, known for its policy optimization capabilities, offers a robust solution by balancing safety, efficiency, and adaptability, particularly in complex situations such as high-speed merging and low-visibility conditions. The simulation results confirm that DQL, A2C, and PPO collectively enhance autonomous vehicle performance by improving navigation capabilities, increasing safety, and adapting more effectively to the complexities of urban traffic environments. This work contributes valuable insights into the application of RL for real-world autonomous driving, providing a detailed comparative evaluation that supports the selection of algorithms tailored to specific driving tasks.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".