Vehicle Localization Using Road Elevation and Inclination Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle localization methods are inefficient in accurately determining the position of a vehicle on non-flat, curved, or distorted road surfaces due to the lack of consideration for road elevation profiles and vehicle inclination, leading to errors in landmark comparison between detected and stored data.
Innovation Solution
The method involves detecting the roadway elevation profile and transforming detected and stored landmarks into a common perspective using parametric models of the roadway and vehicle inclination, allowing for precise comparison and accurate vehicle positioning by minimizing differences between expected and detected profiles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional landmark comparison methods are used without considering elevation profile and vehicle inclination, then the localization process is simpler and faster, but the positioning accuracy deteriorates on non-flat road surfaces
Solution Approach 1:
The patent applies parameter changes by introducing elevation profile parameters and vehicle inclination parameters (pitch and roll angles) to transform landmarks into a common perspective. This allows accurate comparison of landmarks detected by the vehicle's sensors with landmarks stored on digital maps, even on non-flat road surfaces, thereby improving positioning accuracy without requiring additional hardware
Solution Approach 2:
The patent replaces complex mechanical or hardware-based solutions (such as multiple sensor units or physical reference systems) with computational methods. By using parametric models to describe the elevation profile and vehicle inclination, and applying mathematical transformations to landmark coordinates, the system achieves accurate localization through software-based perspective transformation rather than additional physical components
2Measurement precision
If multiple sensor units and data sets are used for detection, then measurement accuracy improves, but computational effort and processing time increase
Solution Approach 1:
The patent applies universality by using a single sensor unit to perform multiple functions: detecting both the elevation profile of the road surface and capturing landmarks for position determination. The same sensor data is processed through parametric models to extract both road geometry information and landmark positions, eliminating the need for separate sensor units and reducing computational overhead while maintaining accuracy
Data Source
AI summary
A method for locating a vehicle involves detecting an elevation profile of a roadway of the vehicle and in which image features on the roadway are detected as landmarks and are compared with landmarks stored on a digital map. A transformation of the detected and/or stored landmarks into a common perspective performed to compare the landmarks. The transformation is carried out based on model parameters of a parametric model of the elevation profile of the roadway and a parametric model of a vehicle inclination. The model parameters are determined by determining an expected elevation profile of the roadway from the parametric models of the elevation profile and the vehicle inclination and minimizing a difference between the expected elevation profile and the detected elevation profile by varying the model parameters.