Vehicle Pose Estimation Using Optimal Filtering and Map Error Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous and semi-autonomous vehicles face challenges in accurately estimating their location relative to features on high-definition digital maps, due to discrepancies between mapped features and their true locations in a global reference frame.
Innovation Solution
The use of optimal filtering processes, such as Kalman or particle filters, in conjunction with satellite positioning systems, inertial measurement units, cameras, and digital maps, allows for the estimation of a vehicle's total pose without correcting the digital map. This involves tracking discrepancies between mapped features and their true locations, and using camera observations to refine pose estimates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If optimal filtering processes are used to estimate vehicle pose without correcting digital maps, then localization accuracy is improved, but computational resources are reduced
Solution Approach 1:
The patent extracts and separates the map correction process from the pose estimation process. Instead of correcting digital map discrepancies, the system extracts only the necessary pose information directly from sensor measurements using optimal filtering, eliminating the computational burden of map correction while maintaining localization accuracy.
Solution Approach 2:
The system uses the vehicle's own sensor measurements (GPS, inertial sensors, wheel encoders) to self-determine its pose without relying on external map correction. The optimal filter processes these self-generated measurements to continuously estimate vehicle position and orientation, making the system self-sufficient and reducing computational complexity.
2Measurement precision
If digital map discrepancies are tracked and corrected, then pose estimation accuracy is improved, but processing time increases
Solution Approach 1:
The system performs preliminary action by pre-defining the state variables and measurement models for the optimal filter before runtime. The filter is configured in advance to process specific sensor measurements (GPS position, inertial data, wheel odometry) directly into pose estimates, eliminating the need for time-consuming map correction operations during actual vehicle operation.
Solution Approach 2:
The patent skips the traditional map correction step entirely in the processing pipeline. Instead of rushing through map alignment and discrepancy correction algorithms, the system directly computes pose from sensor measurements using efficient optimal filtering, significantly reducing processing time while maintaining accuracy.
Data Source
AI summary
A computer includes a processor and a memory storing instructions executable by the processor to determine an estimated pose of a vehicle in a global reference frame having a first error and to determine an estimated second error based on a combination of the first error and a map error, in which the map error represents a difference between the estimated pose of the vehicle in the global reference frame and a corresponding estimated pose in a map-referenced frame. The stored instructions being additionally to predict a third error in a future map-referenced measurement frame based on the estimated second error and a motion model and to compute an update to the third error by combining the predicted third error with an accumulation of instantaneous vehicle position and heading errors obtained via a comparison between a camera-observed feature and a corresponding feature from a digital map.