LiDAR Positioning With Multi-Layer Maps for Overpass Height Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional LiDAR-based positioning solutions for autonomous driving face challenges in complex scenarios with multi-layer road structures, where conventional single-layer Gaussian model maps fail to accurately reflect height information, leading to significant positioning errors and drift.
Innovation Solution
A method that uses a multi-layer single Gaussian model map with point cloud data to determine the estimated position, height, and probability of an object's posture, incorporating a histogram filter for accurate positioning by matching point cloud data with multiple height layers, enabling centimeter-level accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a conventional single-layer Gaussian model map is used for positioning, then the device complexity is reduced, but the positioning precision deteriorates in multi-layer road scenarios
Solution Approach 1:
The map is segmented into multiple layers, each representing a different height level (e.g., ground level, overpass levels). This segmentation allows the system to accurately represent complex multi-layer road structures while maintaining manageable complexity through modular organization of spatial data.
Solution Approach 2:
The positioning system transitions from conventional 2D positioning to 3D positioning by incorporating height information as an additional dimension. This enables accurate positioning in multi-layer road scenarios by matching point cloud data across multiple height layers rather than a single plane.
2Measurement precision
If a multi-layer single Gaussian model map is used, then the positioning precision is improved, but the device complexity increases
Solution Approach 1:
The system changes the parameter representation by using a single Gaussian model to describe the height distribution across multiple layers. This approach maintains mathematical simplicity while capturing complex 3D spatial structures, avoiding the need for overly complex multi-model representations.
Solution Approach 2:
The multi-layer single Gaussian model serves multiple functions: it represents terrain height information, identifies road layers, and enables positioning across different elevation levels. This universal representation reduces the need for separate specialized models for each function.
3Productivity
If conventional positioning methods are used in multi-layer road scenarios, then the processing speed is maintained, but the positioning precision deteriorates due to inability to distinguish height layers
Solution Approach 1:
The system performs preliminary organization of spatial data into multiple height layers before the actual positioning operation. This pre-structuring of data allows for efficient matching during positioning without sacrificing speed, as the multi-layer structure is prepared in advance rather than computed in real-time during positioning.
Data Source
AI summary
The disclosure provides a method, an apparatus, a device and a storage medium for positioning an object. The method includes: obtaining a map related to a region where the object is located, the map including a plurality of map layers having different height information; determining, based on the map and current point cloud data related to the object, an estimated position of the object, an estimated height corresponding to the estimated position and an estimated probability that the object is located at the estimated position with an estimated posture; and determining, at least based on the estimated position, the estimated height and the estimated probability, positioning information for the object, the positioning information indicating at least one of a current position of the object, a current height of the object and a current posture of the object.