Real-time Detection and Avoidance of Obstacles in the Path of Autonomous Vehicles Using Monocular RGB Camera
In this paper, we present an end-to-end real-time detection and collision avoidance framework in an autonomous vehicle using a monocular RGB camera. The proposed system is able to run on embedded hardware in the vehicle to perform real-time detection of small objects. RetinaNet architecture with ResNet50 backbone is used to develop the object detection model using RGB images. A quantized version of the object detection inference model is implemented in the vehicle using NVIDIA Jetson AGX Xavier. A geometric method is used to estimate the distance to the detected object which is forwarded to a MicroAutoBox device that implements the control system of the vehicle and is responsible for maneuvering around the detected objects. The pipeline is implemented on a passenger vehicle and demonstrated in challenging conditions using different obstacles on a predefined set of waypoints. Our results show that the system is capable of detecting objects that appear in an image area as small as 20×30 pixels in a 1280×720 image and can run at a speed of 24 frames per second (FPS) on the embedded device in the vehicle. A data analyzer is also employed to visualize the real-time performance of the system.
- Record URL:
-
Availability:
- Find a library where document is available. Order URL: http://worldcat.org/issn/01487191
-
Supplemental Notes:
- Abstract reprinted with permission of SAE International.
-
Authors:
- Mallik, Apurbaa
- Gaopande, Meghana Laxmidhar
- Singh, Gurjeet
- Ravindran, Aniruddh
- Iqbal, Zafar
- Chao, Steven
- Revalla, Hitha
- Nagasamy, Vijay
-
Conference:
- WCX SAE World Congress Experience
- Location: Detroit & Online Michigan, United States
- Date: 2022-4-5 to 2022-4-7
- Publication Date: 2022-3-29
Language
- English
Media Info
- Media Type: Web
- Features: References;
- Pagination: pp 622-632
-
Serial:
- SAE Technical Paper
- Volume: 5
- Issue Number: 2
- Publisher: Society of Automotive Engineers (SAE)
- ISSN: 0148-7191
- EISSN: 2688-3627
- Serial URL: http://papers.sae.org/
Subject/Index Terms
- TRT Terms: Autonomous vehicles; Computer architecture; Control systems; Crash avoidance systems
- Subject Areas: Highways; Vehicles and Equipment;
Filing Info
- Accession Number: 01841518
- Record Type: Publication
- Source Agency: SAE International
- Report/Paper Numbers: 2022-01-0074
- Files: TRIS, SAE
- Created Date: Apr 6 2022 2:18PM