Advancements and Future Directions of Automotive Radar in Autonomous Vehicles
The advancement of autonomous driving hinges on integrated perception systems combining LiDAR, millimeter-wave radar, and ultrasonic radar. LiDAR employs laser scanning (e.g., 905 nm or 1550 nm wavelengths) to achieve centimeter-level 3D environmental mapping, critical for real-time obstacle detecti...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
EDP Sciences
2025-01-01
|
| Series: | MATEC Web of Conferences |
| Online Access: | https://www.matec-conferences.org/articles/matecconf/pdf/2025/04/matecconf_menec2025_04003.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849411125180891136 |
|---|---|
| author | Chen Zhiyuan |
| author_facet | Chen Zhiyuan |
| author_sort | Chen Zhiyuan |
| collection | DOAJ |
| description | The advancement of autonomous driving hinges on integrated perception systems combining LiDAR, millimeter-wave radar, and ultrasonic radar. LiDAR employs laser scanning (e.g., 905 nm or 1550 nm wavelengths) to achieve centimeter-level 3D environmental mapping, critical for real-time obstacle detection. Millimeter-wave radar (24–77 GHz) provides robust long-range detection (up to 300 meters) and dynamic tracking in adverse weather, while ultrasonic radar enables cost-effective short-range sensing (0.2–5 meters) for parking and low-speed scenarios. Despite their synergy, challenges persist: LiDAR’s susceptibility to weather interference and high costs, millimeter-wave radar’s limited angular resolution, and ultrasonic radar’s range constraints. Additionally, multimodal data fusion (e.g., LiDAR point clouds and radar signals) faces synchronization latency, calibration complexity, and computational demands. Recent innovations include solid-state LiDAR for compact designs, high-frequency millimeter-wave radar (79 GHz) to enhance resolution, and ultrasonic arrays for expanded coverage. Future progress will prioritize AI-driven solutions—such as deep learning for real-time point cloud segmentation and probabilistic classification—alongside vehicle-to-infrastructure (V2X) collaboration. These strategies aim to optimize sensor synergy, reduce costs, and improve reliability, accelerating the commercialization of SAE Level 4/5 autonomous vehicles and enabling intelligent transportation networks focused on safety and scalability. |
| format | Article |
| id | doaj-art-edc50738b9bd4475a15433fb4058a039 |
| institution | Kabale University |
| issn | 2261-236X |
| language | English |
| publishDate | 2025-01-01 |
| publisher | EDP Sciences |
| record_format | Article |
| series | MATEC Web of Conferences |
| spelling | doaj-art-edc50738b9bd4475a15433fb4058a0392025-08-20T03:34:52ZengEDP SciencesMATEC Web of Conferences2261-236X2025-01-014100400310.1051/matecconf/202541004003matecconf_menec2025_04003Advancements and Future Directions of Automotive Radar in Autonomous VehiclesChen Zhiyuan0Institut Hohai-Lille, Hohai UniversityThe advancement of autonomous driving hinges on integrated perception systems combining LiDAR, millimeter-wave radar, and ultrasonic radar. LiDAR employs laser scanning (e.g., 905 nm or 1550 nm wavelengths) to achieve centimeter-level 3D environmental mapping, critical for real-time obstacle detection. Millimeter-wave radar (24–77 GHz) provides robust long-range detection (up to 300 meters) and dynamic tracking in adverse weather, while ultrasonic radar enables cost-effective short-range sensing (0.2–5 meters) for parking and low-speed scenarios. Despite their synergy, challenges persist: LiDAR’s susceptibility to weather interference and high costs, millimeter-wave radar’s limited angular resolution, and ultrasonic radar’s range constraints. Additionally, multimodal data fusion (e.g., LiDAR point clouds and radar signals) faces synchronization latency, calibration complexity, and computational demands. Recent innovations include solid-state LiDAR for compact designs, high-frequency millimeter-wave radar (79 GHz) to enhance resolution, and ultrasonic arrays for expanded coverage. Future progress will prioritize AI-driven solutions—such as deep learning for real-time point cloud segmentation and probabilistic classification—alongside vehicle-to-infrastructure (V2X) collaboration. These strategies aim to optimize sensor synergy, reduce costs, and improve reliability, accelerating the commercialization of SAE Level 4/5 autonomous vehicles and enabling intelligent transportation networks focused on safety and scalability.https://www.matec-conferences.org/articles/matecconf/pdf/2025/04/matecconf_menec2025_04003.pdf |
| spellingShingle | Chen Zhiyuan Advancements and Future Directions of Automotive Radar in Autonomous Vehicles MATEC Web of Conferences |
| title | Advancements and Future Directions of Automotive Radar in Autonomous Vehicles |
| title_full | Advancements and Future Directions of Automotive Radar in Autonomous Vehicles |
| title_fullStr | Advancements and Future Directions of Automotive Radar in Autonomous Vehicles |
| title_full_unstemmed | Advancements and Future Directions of Automotive Radar in Autonomous Vehicles |
| title_short | Advancements and Future Directions of Automotive Radar in Autonomous Vehicles |
| title_sort | advancements and future directions of automotive radar in autonomous vehicles |
| url | https://www.matec-conferences.org/articles/matecconf/pdf/2025/04/matecconf_menec2025_04003.pdf |
| work_keys_str_mv | AT chenzhiyuan advancementsandfuturedirectionsofautomotiveradarinautonomousvehicles |