CarVision: Estimating relative speed and distance with mmWave Radar
Poor visibility in adverse weather conditions, like dense fog or heavy rain, significantly increases the risk of rear-end collisions. So it is important to maintain a safer distance from the vehicle ahead. CarVision does the same ensuring safer driving experience.
Key Features
- Robust Performence: CarVision utilizes mmWave radars, which gives robust performence in any adverse weather conditions. Also not affected by airborn particles like dust, smoke, etc.
- Cost-Efficient & Scalable: Unlike expensive sensors used in ADAS like LiDAR, our system offers a highly deployable and affordable solution without compromising accuracy.
- Comprehensive Monitoring: CarVison can continiously monitor the front vehicle, with a wide range of variety of vehicle segment in Idian roads. For example, Hatchback, Sedan, and SUV.
- Realworld Data: In total, 15 hours of driving data was collected for a duration of 5 days. The data comprises diverse driving environments, including 4 hours on college campus roads, 3 hours on urban and rural roads, and 8 hours on highways both day and night time.
- Alert System: CarVision can alert the driver through android application in case of front vehicle is too close to the rear vehicle or the front vehicle has a safer distance with respect to the rear vehicle.
- Energy Efficient: CarVision can be easily deployed on a low-compute edge device like Raspberry Pi-4 for live inference with an average inference time of 0.57 seconds.
Licensing:The platform source and the collected dataset is free to download and can be used with GNU Affero General Public License for non-commercial purposes. All participants signed forms consenting to the collected mmwave dataset and activity labels for non-commercial research purposes. The institute's ethical review committee has approved the field study (Order No: IIT/SRIC/DEAN/2023, Dated July 31, 2023).
Contributors
Rajib Sarkar
IIT Kharagpur, India
Argha Sen
IIT Kharagpur, India
Sandip Chakraborty
IIT Kharagpur, India
Teaser Video
Publications
- Rajib Sarkar, Argha Sen, and Sandip Chakraborty: "CarVision: Vehicle Ranging and Tracking Using mmWave Radar for Enhanced Driver Safety", IEEE PerCom 2025.