Check patentability & draft patents in minutes with Patsnap Eureka AI!

Partially observable driving planning method based on adaptive particles and belief filling

An adaptive and automatic driving technology, applied in two-dimensional position/course control, instruments, control/regulation systems, etc., can solve problems such as small weight, inequality, and weight degradation

Active Publication Date: 2021-07-30
NANJING UNIV
View PDF6 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This mode is called sequential importance sampling, which has a flaw, that is, after several rounds of particle propagation, the inequality between the weights of each particle will eventually lead to weight degradation, that is, most Weights are concentrated on a small number of particles, while all other particles have negligibly small weights

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
  • Partially observable driving planning method based on adaptive particles and belief filling
  • Partially observable driving planning method based on adaptive particles and belief filling
  • Partially observable driving planning method based on adaptive particles and belief filling

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0066] In this embodiment, based on the KLD-Sampling method, the lane-changing task on a multi-lane road is solved online using an automatic driving online planning method, including:

[0067] When performing a lane change task on a multi-lane road, since the road type and other information have not changed, the number of lanes and the width of the lane can be simply used as road information, and they will not change during the entire task.

[0068] Vehicles on the road are identified using sensor observations and modeled as described previously, specifically considering a scenario involving multiple vehicles. The physical state of the vehicle itself is described by (x, y, θ, v, a, l, w), as shown in the attached figure 2 As shown, where x, y represent the coordinates of the vehicle, θ represents the counterclockwise deflection angle between the front of the vehicle and the positive direction of the x-axis, v, a represent the velocity and acceleration, and l, w represent the ...

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 a partially observable driving planning method based on adaptive particles and belief filling, and the method comprises the steps: modeling an automatic driving task into a partially observable Markov decision task, and carrying out the real-time solving of the partially observable Markov decision task through an efficient online planning solving method. Due to the adoption of the online solving method, the system can support flexible modeling solving of various road types, obstacle types and intelligent agent types, and is a universal intelligent driving planning method. According to the online planning solving method adopted by the method, efficient approximation of a belief state is realized by utilizing adaptive particle filtering, and a belief filling method is introduced to merge similar observation branches, so that tasks with huge observation space, such as automatic driving, can be efficiently solved.

Description

technical field [0001] The invention relates to a partially observable driving planning method based on adaptive particles and belief filling, belonging to the technical field of automatic driving. Background technique [0002] The automatic driving task is a typical partially observable task. On the one hand, the occlusion caused by obstacles makes the observation in the automatic driving task incomplete. On the other hand, the intention and driving style of other agents are naturally partially observable states. It needs to be speculated based on observation, and the incompleteness of this observation brings about the uncertainty of the state. However, the current methods such as finite state machine, decision tree model and other methods cannot deal with this uncertainty well. [0003] Partially observable Markov decision process is an extension of Markov decision process (MDP), which additionally considers the uncertainty of the state. In MDP, the agent can accurately ...

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
IPC IPC(8): G05D1/02
CPCG05D1/0214G05D1/0221G05D1/0276Y02T10/40
Inventor 章宗长俞扬周志华吴晨阳杨国钰
Owner NANJING UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com