Radar-IMU Map Feedback for Urban Vehicle Positioning Accuracy

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Traditional self-driving vehicle positioning systems, relying on IMU and GNSS, face challenges in achieving high accuracy and reliability due to errors in IMU-based systems and limitations in GNSS performance, especially in dense urban areas where satellite visibility is obstructed, leading to reduced positioning accuracy and availability.

Innovation Solution

Integration of radar measurements with motion sensor data using nonlinear state estimation techniques to generate an integrated navigation solution, which updates the navigation system and provides feedback for improving map information accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional IMU/GNSS integration is used for positioning, then the system can provide continuous positioning data, but positioning accuracy deteriorates in dense urban areas due to satellite blockage and multipath effects

Engineering Contradiction:
Improvepositioning availabilityVSAvoidpositioning accuracy
Core Design Contradiction:
ReliabilityVSMeasurement precision

Solution Approach 1:

The patent combines radar measurements with IMU data through a nonlinear state estimation technique, merging two different sensing modalities (radio wave-based radar and motion-based IMU) to create an integrated navigation solution that compensates for the weaknesses of each individual system in urban environments

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent introduces map information as an intermediary element that mediates between radar measurements and positioning determination. The map provides prior knowledge about the environment that helps interpret radar data and constrain possible positions, improving accuracy without requiring direct satellite signals

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If standalone IMU-based positioning is used, then the system provides accurate short-term relative pose estimation, but error accumulates exponentially over time due to drift

Engineering Contradiction:
Improveshort-term positioning accuracyVSAvoidlong-term positioning stability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent implements a feedback mechanism where radar measurements and map information continuously constrain and correct the IMU-based position estimates. The nonlinear state estimation technique uses radar-derived range and bearing information to feedback-correct the drifting IMU solution, preventing error accumulation over time

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent uses map information as preliminary knowledge before determining the current position. The pre-stored map data provides expected environmental features and geometries that are used in advance to constrain and guide the interpretation of real-time radar measurements, preventing drift accumulation

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If GNSS signals are used for absolute positioning, then the system provides stable long-term position data, but signal availability is reduced in dense urban areas due to blockage and multipath

Engineering Contradiction:
Improveabsolute positioning accuracyVSAvoidsignal availability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent makes the navigation system universal by enabling it to function accurately in multiple environments (open areas and dense urban canyons) using the same integrated radar-IMU-map approach. This multi-functional solution replaces the environment-dependent GNSS system with a sensor fusion approach that works universally across different urban and non-urban settings

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent substitutes the mechanical/electromagnetic GNSS signal reception system with a radar-based active sensing system. Instead of passively receiving satellite signals that are blocked by buildings, the system actively emits radar waves and processes reflections, replacing the GNSS dependency with a radar-IMU fusion approach that functions independently of satellite visibility

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

4Device complexity

If conventional map information is used for navigation aid, then the system can constrain position to possible paths, but map inaccuracies and lack of detail reduce positioning precision

Engineering Contradiction:
Improvenavigation constraint simplicityVSAvoidpositioning precision
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent dynamically updates and refines map information using feedback from the integrated navigation solution. As the vehicle traverses the environment, radar measurements and determined positions are used to update the map data, making it progressively more accurate and detailed for future navigation constraints

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent uses pre-collected map information as preliminary data to initialize the navigation system and provide initial constraints. The map serves as prior knowledge that guides the interpretation of radar measurements before real-time updates occur, enabling the system to function even before the map is fully refined

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3904908A1Method and system for map improvement using feedback from positioning based on radar and motion sensors
Publication Date: 2021.11.03 TRUSTED POSITIONING
  • EP3904908A1 patent drawingFigure 1
  • EP3904908A1 patent drawingFigure 2
  • EP3904908A1 patent drawingFigure 3

AI summary

Feedback for map information is based on an integrated navigation solution for a device within a moving platform using obtained motion sensor data from a sensor assembly of the device, obtained radar measurements for the platform and obtained map information for an environment encompassing the platform. An integrated navigation solution is generated based at least in part on the obtained motion sensor data using a nonlinear state estimation technique that uses a nonlinear measurement model for radar measurements. The map information is assessed based at least in part on the integrated navigation solution and radar measurements so that feedback for the map information can be provided.