Weighted Traffic Stream Fitting for Congested Lane Inference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current intelligent driving systems face challenges in accurately identifying lane lines, especially in congested traffic conditions or due to unclear or outdated lane markings, which affects vehicle lateral positioning and path planning.
Innovation Solution
The method involves determining traffic stream information by grouping target vehicles and calculating fitting weights based on their motion and state information, allowing for the generation of current traffic stream information that can replace lane line data for vehicle positioning and control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If lane line recognition is used for vehicle lateral positioning, then the system can determine vehicle position and travel path, but the accuracy deteriorates in congested traffic conditions with dense traffic streams and unclear lane lines
Solution Approach 1:
The patent introduces traffic stream information as an intermediary element to bridge the gap between vehicle positioning requirements and lane line recognition limitations. By extracting trajectory information from surrounding vehicles' motion patterns, the system creates an alternative reference system that mediates the positioning function when traditional lane line recognition fails in congested environments
Solution Approach 2:
The patent replaces the optical-mechanical lane line recognition system with a computational traffic stream analysis system. Instead of relying on camera-based visual detection of physical lane markings, the system substitutes a algorithmic approach that processes vehicle trajectory data to infer positional information, thereby overcoming the limitations of mechanical/optical recognition in dense traffic
2Loss of information
If forward camera sensing is used to collect image data for lane line identification, then the system can detect lane markings, but the sensing distance and field of view are limited in congested traffic conditions
Solution Approach 1:
The patent merges multiple information sources to compensate for the limited camera field of view. By combining trajectory data from multiple surrounding vehicles detected within the constrained visual range, the system synthesizes a broader spatial understanding that exceeds the physical limitations of the camera's sensing area
Solution Approach 2:
The patent transitions from two-dimensional image plane analysis to three-dimensional spatial trajectory analysis. By utilizing the temporal dimension of vehicle motion trajectories, the system extracts positional and directional information that compensates for the limited spatial coverage of the camera, effectively adding a temporal dimension to overcome spatial constraints
3Measurement precision
If traffic stream information is used instead of lane line recognition, then the accuracy of vehicle lateral positioning is enhanced in congested environments, but the system complexity increases
Solution Approach 1:
The patent makes the traffic stream analysis module multi-functional, enabling it to serve multiple purposes: vehicle positioning, lane inference, and traffic pattern recognition. This universal approach allows the same computational framework to address various positioning challenges without requiring separate specialized systems for each function
Solution Approach 2:
The traffic stream information system is self-sustaining, using the motion trajectories of surrounding vehicles as both the input data and the reference framework for positioning. The system automatically adapts to changing traffic conditions by continuously updating its model of vehicle flow patterns, eliminating the need for external calibration or manual intervention
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
A method and a device for determining traffic stream information, electronic equipment and a storage medium are provided.The method comprises: grouping each of target vehicles based on motion information of a host vehicle and state information of one or more of the target vehicles to obtain grouping information (401); determining a fitting weight for each of the target vehicles based on the state information and the grouping information for each of the target vehicles (402); and generating, by the fitting, one or more pieces of current traffic stream information based on the motion information of the host vehicle, the state information of each of the target vehicles, the fitting weights and the grouping information of each of the target vehicles (403). By grouping a plurality of target vehicles and determining the fitting weight of each of the target vehicles, the method can generate the current traffic stream information by fitting based on the grouping information, the fitting weight, the motion information of the host vehicle and the state information of the target vehicle, so as to achieve the sensing of traffic stream.