In this paper, an innovative fusion scheme is proposed, which uses Faster R-CNN (Faster Region Convolutional Neural Network) for environment perception, and combines DDPG (Deep Deterministic Policy Gradient) algorithm to optimize the intelligent chassis wire-by-wire technology. Firstly, Faster R-CNN is used as the front-end perception module to realize the real-time detection and classification of complex targets in road environment. The model can quickly and accurately generate the bounding box and category label of the target, which provides basic data for subsequent intelligent decision-making. Then, the DDPG algorithm is used as the decision-making and control core. Based on the real-time perception information provided by Faster R-CNN, DDPG can learn and execute a series of complex control actions to ensure the safe and efficient driving of the vehicle. By optimizing the two key links of perception and decision-making, the autonomous driving ability and safety of autonomous driving vehicles are significantly improved, so that the vehicles can continuously optimize their behavior in complex driving scenarios and deal with various driving situations.
Research on Intelligent Chassis Drive-by-Wire Technology Using Computer Vision in Autonomous Driving
27.09.2024
1206380 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Intelligent networked automobile drive-by-wire chassis
Europäisches Patentamt | 2024
|Intelligent automobile drive-by-wire chassis control system
Europäisches Patentamt | 2023
|Intelligent drive-by-wire chassis comprehensive control platform
Europäisches Patentamt | 2024
|