IMU Calibration Using Vehicle Localization Data

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional maps lack the precision and accuracy required for safe navigation of autonomous vehicles, with GPS systems providing inaccuracies and traditional mapping methods being expensive, time-consuming, and unable to keep up with frequent road updates.

Innovation Solution

The use of high-definition maps that are maintained and updated through data collected by autonomous vehicles themselves, allowing for precise location determination and real-time navigation, with inertial measurement unit calibration to enhance accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional maps are used for autonomous vehicle navigation, then the system can operate with simpler infrastructure, but the location accuracy and precision fall short of the required 30 cm threshold

Engineering Contradiction:

Improvelocation accuracyVSAvoidmapping system complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The autonomous vehicles perform self-calibration by using their own sensor data (LIDAR, cameras, IMU) to generate and update HD maps. The vehicles localize themselves relative to the map and use this information to calibrate their IMUs, eliminating the need for external survey teams and specialized survey cars.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system continuously refines map accuracy by comparing vehicle localization results against the HD map and using the discrepancies to calibrate IMUs. This feedback loop ensures that location precision improves over time as more vehicles contribute data and corrections are applied.

Inventive Principle:

Principle #23Feedback

2Productivity

If traditional survey teams with specialized cars create maps, then initial map creation is possible, but the process is expensive and time-consuming (weeks to months)

Engineering Contradiction:

Improvemap creation speedVSAvoidmap update latency

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

Autonomous vehicles perform self-calibration by using their own sensor data (LIDAR, cameras, IMU) to generate and update HD maps. The vehicles localize themselves relative to the map and use this information to calibrate their IMUs, eliminating the need for external survey teams and specialized survey cars.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

Instead of periodic updates by survey teams, the HD map is continuously updated as autonomous vehicles traverse the environment. Each vehicle contributes real-time calibration data, ensuring the map remains current with recent road changes and modifications.

Inventive Principle:

Principle #20Continuity of useful action

3Measurement precision

If GPS systems are used for location determination, then the system can operate with minimal infrastructure, but the accuracy is insufficient (3-5 meters to over 100 meters)

Engineering Contradiction:

Improvelocation accuracyVSAvoidlocalization system complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The HD map serves as an intermediary reference framework that bridges GPS coarse location data and precise vehicle localization. Vehicles use the map to interpret GPS signals and sensor data, achieving meter-level or sub-meter accuracy through map-matching and sensor fusion techniques.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

Inventive Principle:

Principle #23Feedback

4Adaptability or versatility

If survey fleets with many cars are deployed to cover all roads, then map completeness can be improved, but the cost and time required increase significantly

Engineering Contradiction:

Improvemap coverageVSAvoidsurvey fleet complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

Any autonomous vehicle equipped with standard sensors (LIDAR, cameras, IMU) can contribute to map creation and maintenance. The system is universally applicable across different vehicle types, allowing a diverse fleet of autonomous vehicles to collectively maintain comprehensive and up-to-date maps.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS11835342B2Calibration of inertial measurement units of vehicles using localization

Publication Date: 2023.12.05 NVIDIA CORP

AI summary

Operations of the present disclosure include obtaining a first measure of velocity of a vehicle based on a plurality of locations determined for the vehicle. The operations also include obtaining, based on IMU measurements of an inertial measurement unit (IMU) of the vehicle, a second measure of velocity of the vehicle. In addition, the operations include performing calibration of the IMU based on the first measure of velocity and the second measure of velocity.