Unlock instant, AI-driven research and patent intelligence for your innovation.

Embedded optimized control method suitable for unmanned driving and its driving control module and automatic driving control system

A control method and unmanned driving technology, applied in control devices, neural learning methods, biological neural network models, etc., can solve the problems of easy sudden changes and difficult to accurately predict the behavior of other vehicles, and achieve the effect of improving the efficiency of the algorithm

Active Publication Date: 2022-04-19
吉林大学青岛汽车研究院
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] For example, the driving decision-making problem of unmanned vehicles in the field of vehicle control where people and vehicles are mixed, because the behavior of other vehicles is difficult to predict accurately and is prone to sudden changes
Therefore, there are often uncertain factors in the environment, which are difficult to be accurately predicted in advance

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Embedded optimized control method suitable for unmanned driving and its driving control module and automatic driving control system
  • Embedded optimized control method suitable for unmanned driving and its driving control module and automatic driving control system
  • Embedded optimized control method suitable for unmanned driving and its driving control module and automatic driving control system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0066] An embedded optimized control method suitable for unmanned driving, the control method includes the following steps:

[0067] Step 1. Establishment of driving decision-making problem and characterization of driving strategy;

[0068] Based on the reinforcement learning method, the driving decision is modeled as a Markov decision process. The driving decision based on the Markov decision process includes the state vector S representing the relative state of the vehicle and surrounding vehicles; and the action vector A representing the driving decision of the vehicle based on the parameterized driving decision framework. The action vector A contains discrete decision variables, the lateral offset T of the terminal relative to the centerline of the lane y , pointing to driving behaviors such as lane keeping, left lane changing, right lane changing, etc., and a continuous decision variable, expected acceleration a tar , action time t a . So the action vector A=(T y ,a ...

Embodiment 2

[0110] A driving control module, the driving control module includes a computer installed with the algorithm program of the embedded optimization control method suitable for unmanned driving described in Embodiment 1.

Embodiment 3

[0112] An automatic driving control system, the automatic driving control system includes a perception and cognition module, the driving control module described in Embodiment 2, and a trajectory planning module.

[0113] Each sub-control system of the automatic driving control system of an unmanned vehicle needs to realize automatic control through system design. Such as figure 1 As shown, it includes perception and cognition module, driving control module and trajectory planning module. The method of Embodiment 1 is mainly aimed at the driving control module.

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

No PUM Login to View More

Abstract

The invention discloses an embedded optimization control method suitable for unmanned driving, its driving control module and an automatic driving control system. The learning method includes the establishment of driving decision-making problems and the characterization of driving strategies; the establishment of neural network experience models; continuous driving decision-making Variable solving; discrete driving decision variable learning; it is based on the needs of the above practical problems. For control problems involving continuous control that are difficult to obtain in advance with variable application scenarios, model-based methods can be used to effectively search the action space in reinforcement learning , and enable rapid iteration of the driving strategy.

Description

technical field [0001] The invention relates to the technical field of unmanned driving, in particular to an embedded optimization control method suitable for unmanned driving, a driving control module thereof, and an automatic driving control system. Background technique [0002] With the continuous development of reinforcement learning technology, it is more and more applied to various problems. Therefore, for different control problems, further requirements are put forward for the algorithm efficiency of reinforcement learning in practical applications. In the field of control, for continuous control problems, in general, model-free reinforcement learning algorithms lack high algorithm efficiency and algorithm stability. In order to improve the efficiency and stability of the algorithm, some professional methods and technologies in the field of reinforcement learning have been proposed, for example, asynchronous update strategy, pre-training method, return integer techno...

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
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): B60W60/00B60W50/00G06N3/04G06N3/08
CPCB60W60/001B60W50/00G06N3/086B60W2050/0003G06N3/044
Inventor 张羽翔王玉海丛岩峰高炳钊陈虹
Owner 吉林大学青岛汽车研究院