Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
AUTOMATIC PARKING DECISION MAKING METHOD BASED ON FUSION OF MODEL PREDICTIVE CONTROL AND REINFORCEMENT LEARNING
Document Type and Number:
WIPO Patent Application WO/2024/088068
Kind Code:
A1
Abstract:
Provided is an automatic parking decision making method based on the fusion of model predictive control and reinforcement learning, relating to the technical field of autonomous driving. In the process of automatic parking, the process of selecting an optimal action is trained by using a proximal policy optimization algorithm, wherein the proximal policy optimization algorithm comprises an Actor network and a Critic network, a policy function corresponds to the Actor network, a value function corresponds to the Critic network, the output of the Actor network is the probability of taking each possible steering action when a vehicle is in a specific state, and the output of the Critic network is a state value function of the state after the steering action of the Actor network is taken. In the context of narrow and vertical parking spaces, model predictive control is combined with a proximal policy optimization reinforcement learning algorithm, so that the problems that the parking planning time is long and the success rate is low under the condition that the parking spaces are narrow are solved, and the parking space selection range of the system is expanded.

Inventors:
SHI JUNREN (CN)
LI KEXIN (CN)
CHEN MAOXING (CN)
PU CHANGHAO (CN)
HE JIA (CN)
SU YONGKANG (CN)
LIU MINGJIE (CN)
CHEN JUNSHENG (CN)
HUANG JIAN (CN)
Application Number:
PCT/CN2023/124189
Publication Date:
May 02, 2024
Filing Date:
October 12, 2023
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
UNIV CHONGQING POSTS & TELECOM (CN)
International Classes:
G05B19/042; B60W30/06
Foreign References:
CN112158196A2021-01-01
KR102457914B12022-10-24
CN108334086A2018-07-27
CN115657548A2023-01-31
CN109492763A2019-03-19
CN112419775A2021-02-26
CN114454875A2022-05-10
US20220092973A12022-03-24
Attorney, Agent or Firm:
TIDYTEND INTELLECTUAL PROPERTY LAW FIRM (CN)
Download PDF: