Lane-Level Map Generation Using Particle Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for generating accurate lane-level maps for autonomous driving systems require significant manual effort and are inefficient, as they often rely on tedious manual annotation or extensive data collection using GPS tracks.
Innovation Solution
A method that uses Lidar data to determine road lane characteristics by applying particle filtering techniques, including regular, dual, and mixture particle filters, in conjunction with GraphSlam algorithms, to accurately estimate lane information such as number, width, and center, reducing the need for manual annotation and improving data collection efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual annotation or GPS track collection is used to generate lane-level maps, then map accuracy can be achieved, but significant manual effort and time consumption occur
Solution Approach 1:
The system enables autonomous vehicles to self-map lane-level information by utilizing their own sensor data (Lidar, cameras, GPS) and processing it through particle filtering algorithms. The vehicle serves its dual purpose of navigation and map generation, eliminating the need for dedicated manual annotation or data collection operations.
Solution Approach 2:
The patent replaces manual mechanical processes (physical driving, manual annotation) with automated computational processes. Particle filtering algorithms automatically process sensor data to extract lane characteristics, substituting human labor with algorithmic processing that achieves comparable or superior accuracy.
2Extent of automation
If automated driving systems use traditional map generation methods, then existing infrastructure can be utilized, but the systems cannot achieve the required high-precision lane-level information
Solution Approach 1:
The system dynamically adapts to different road conditions and configurations in real-time using particle filtering. Rather than relying on static pre-mapped data, the system continuously updates lane representations based on current sensor observations, enabling high-precision mapping across diverse and changing environments.
Solution Approach 2:
The patent transitions from traditional 2D map representations to 3D spatial understanding by incorporating Lidar depth information and vertical lane characteristics. This dimensional expansion enables the system to capture complex lane configurations, intersections, and spatial relationships that traditional flat maps cannot represent.
3Manufacturing precision
If particle filtering and Lidar data processing are applied, then lane characteristics can be accurately estimated, but computational complexity increases
Solution Approach 1:
The system segments the complex task of lane mapping into manageable components: Lidar data processing, GPS positioning, particle filtering for lane estimation, and GraphSlam for map integration. Each module handles a specific aspect independently, making the overall complex system more manageable and easier to optimize.
Solution Approach 2:
The system performs preliminary data processing and feature extraction before applying the computationally intensive particle filtering. By pre-processing Lidar data to identify potential lane markers and extracting key geometric features in advance, the system reduces the computational burden during the main mapping operation.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This method enables the generation of accurate lane-level maps with reduced preparation effort, effectively capturing complex lane configurations like bicycle lanes and transitions, improving the precision and efficiency of lane information extraction for autonomous vehicle navigation.
Implementation Method 1
obtaining Lidar data of road markers on a road
Implementation Method 2
a rotating laser beam and a reflected laser beam detector
Implementation Method 3
detecting the intensity of the laser beam reflected from reflective road markers
Data Source
AI summary
A method for generating accurate lane level maps based on course map information and Lidar data captured during the pass of a sensor carrying vehicle along a road. The method generates accurate lane estimates including the center of each lane, the number of lanes, and the presence of any bicycle paths and entrance and exit ramps using a computer-implemented method where the course map data and the Lidar data are subjected to particle filtering.