Autonomous driving behavior optimization method, device and computer equipment for unmanned vehicles
A technology for autonomous driving and unmanned vehicles, which is applied to computer components, calculations, character and pattern recognition, etc., and can solve problems such as low learning efficiency, poor scene migration and large errors in the autonomous driving system of unmanned vehicles
Active Publication Date: 2021-09-10
NAT UNIV OF DEFENSE TECH
View PDF10 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] Based on this, it is necessary to address the above technical problems and provide a method, device, computer equipment and storage medium that can solve the problems of poor scene mobility, large errors, and low learning efficiency in the autonomous driving system of unmanned vehicles
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0050] In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present application, and are not intended to limit the present application.
[0051] The self-driving behavior optimization method for unmanned vehicles provided in this application can be applied in the following application environments. The first scene is defined as the actual driving scene of the vehicle, and the second scene is the pre-training scene of the vehicle behavior early warning model. After the vehicle is trained in the early warning model in the second scene, it is driven in the first scene. The decision-making behavior of controlling the vehicle is obtained by the pre-trained autonomous vehicle autonomous driving model according to t...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The present application relates to a method, device and computer equipment for optimizing autonomous driving behavior of unmanned vehicles. The method includes: acquiring vehicle driving status information, inputting a pre-trained autonomous driving model to obtain decision-making behavior information for controlling vehicle driving, acquiring a first image sequence of vehicle driving, and inputting a pre-trained vehicle behavior early warning model to obtain a risk assessment value. Obtain user feedback information through human-machine collaboration, and judge whether the risk assessment value output by the vehicle behavior early warning model is accurate according to the feedback information. When the risk assessment value is inaccurate, regenerate a new tag value and correct the vehicle behavior early warning model; When the risk assessment value is accurate, output the immediate reward value for the vehicle's driving behavior. According to the vehicle's state information, decision-making behavior information and the immediate reward value of the vehicle under the first image sequence, the autonomous driving model is optimized for training. The autonomous driving model for vehicle driving behavior decision-making.
Description
technical field [0001] The present application relates to the field of unmanned driving technology, in particular to a method, device and computer equipment for optimizing autonomous driving behavior of unmanned vehicles. Background technique [0002] Deep reinforcement learning is a very representative business card of today's artificial intelligence technology. The computer Go programs AlphaGo and Alpha Zero with deep reinforcement learning as the core have been able to completely crush the highest level of human professional players. Although deep reinforcement learning has shown great potential in many aspects, the application of deep reinforcement learning in unmanned driving is not outstanding compared with other fields. [0003] The reason for this situation is that the training process of deep reinforcement learning is essentially a trial and error process. In the training phase, it needs to collect various decision-making behaviors that the unmanned vehicle may ma...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/00G06K9/34G06K9/62G06N3/04G06Q10/06
CPCG06Q10/0635G06V20/54G06V20/56G06V10/267G06N3/045G06F18/214
Inventor 史美萍吴涛陈杰刘大学
Owner NAT UNIV OF DEFENSE TECH