Collision avoidance strategies in autonomous vehicles and on-ramp scenario: A review
Collision avoidance (CA) in autonomous vehicles (AVs) is essential for the safety and efficiency of modern transportation systems. This paper delves into various strategies and methodologies for CA, categorizing them to improve clarity and comprehension. The research primarily reviews peer-reviewed...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier Ltd
2025
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/43869/ http://umpir.ump.edu.my/id/eprint/43869/1/Pulished%20Journal%20Fuad.pdf |
| Summary: | Collision avoidance (CA) in autonomous vehicles (AVs) is essential for the safety and efficiency of modern transportation systems. This paper delves into various strategies and methodologies for CA, categorizing them to improve clarity and comprehension. The research primarily reviews peer-reviewed journals and conference proceedings from the past five years, though notable older studies are also considered. Non-ground AVs research was excluded from this analysis. The CA strategies identified are grouped into six categories: combination of path planning and path tracking control (PP + PTC), path planning (PP), steering, braking, combination of steering and braking, and other methods. Among these, the PP + PTC strategy was the most common, used in 44 cases (38.9%), followed by PP in 16 cases (14.2%), steering in 15 cases (13.3%), other methods and combination of steering and braking in 13 cases each (11.5%), and braking in 12 cases (10.6%). Additionally, the study highlights the on-ramp scenario as an area needing more research. For this scenario, connected AVs (CAV) was the most frequently studied strategy, with 11 cases, followed by machine learning approaches with 9 cases, and other methods with 3 cases. The results underscore the importance of the PP + PTC strategy for effective CA, as it combines PP with PTC to execute planned trajectories efficiently. These insights aim to aid in developing more robust and reliable CA systems in AVs, contributing to safer and more efficient transportation. |
|---|