Edge Device Deployment of Multi-Tasking Network for Self-Driving Operations
This addresses the problem of efficient perception for autonomous driving on resource-constrained edge devices, but it is incremental as it applies existing multi-tasking methods to a specific deployment scenario.
The paper tackled deploying a multi-tasking network for object detection, drivable area segmentation, and lane detection on an embedded system (Nvidia Jetson Xavier NX) for self-driving operations, achieving deployment on edge devices with comparisons based on different backbone networks.
A safe and robust autonomous driving system relies on accurate perception of the environment for application-oriented scenarios. This paper proposes deployment of the three most crucial tasks (i.e., object detection, drivable area segmentation and lane detection tasks) on embedded system for self-driving operations. To achieve this research objective, multi-tasking network is utilized with a simple encoder-decoder architecture. Comprehensive and extensive comparisons for two models based on different backbone networks are performed. All training experiments are performed on server while Nvidia Jetson Xavier NX is chosen as deployment device.