Vehicle Heading Estimation Using Doppler-Constrained Scan Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in accurately estimating the heading of target vehicles in real-time, which is crucial for navigation and behavior inference, due to limitations in existing sensor data processing and comparison methods.
Innovation Solution
A method using a modified iterative closest point (ICP) algorithm, combined with Doppler speed radar estimates, to compare and align point clouds from initial and subsequent laser scans, allowing for precise determination of the target vehicle's heading by minimizing distance between corresponding points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional sensor data processing methods are used for heading estimation, then the system complexity is low, but the measurement precision of heading is insufficient
Solution Approach 1:
The patent combines multiple sensor data sources (laser scanner point clouds and Doppler speed radar) into a unified processing framework. The ICP algorithm integrates geometric information from point cloud comparisons with velocity information from Doppler radar to jointly estimate heading, achieving higher precision than individual sensors could provide alone.
Solution Approach 2:
The system performs preliminary actions by using Doppler speed radar to estimate velocity before the main heading estimation process. This pre-estimated velocity serves as a constraint and initialization for the ICP algorithm, improving convergence speed and accuracy while reducing the computational burden during real-time processing.
2Measurement precision
If real-time heading estimation is performed using detailed point cloud comparison, then the measurement precision improves, but the processing time increases
Solution Approach 1:
The system performs preliminary velocity estimation using Doppler radar before executing the computationally intensive ICP algorithm. This pre-processing step provides initial constraints that guide the point cloud comparison, enabling faster convergence and reducing the time required for detailed geometric analysis.
Solution Approach 2:
The system implements feedback by using the velocity information from Doppler radar to constrain and guide the ICP optimization process. This feedback mechanism allows the algorithm to focus computational efforts on relevant transformations, improving efficiency while maintaining precision in real-time 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 approach enables accurate and efficient estimation of a target vehicle's heading, enhancing the autonomous vehicle's ability to navigate and react to other vehicles' behavior in real-time, improving safety and operational efficiency.
Implementation Method 1
combined with Doppler speed radar estimates
Implementation Method 2
point clouds from initial and subsequent laser scans
Data Source
AI summary
Methods and devices for estimating a heading of a target vehicle are disclosed. An example method may include determining a first point cloud representative of a location of a target vehicle at a first time period and a second point cloud representative of the location of the target vehicle at a second time period. Using a computing device, an initial comparison between the first point cloud and the second point cloud may be determined based on an estimate of a speed for the target vehicle and a time difference between the first time period and the second time period. Additionally, the initial comparison may be revised based on a minimization of a distance between points of the first point cloud and corresponding points of the second point cloud. An estimate of a heading of the target vehicle may then be determined based on the revised comparison.