Integrated Navigation Using Nonlinear State Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional navigation systems for autonomous vehicles, such as IMU/GNSS integration, face challenges in achieving accurate and reliable pose estimation, especially in dense urban areas where GNSS signals are blocked or affected by multipath, leading to reduced positioning accuracy.
Innovation Solution
Integrating optical sensor data with motion sensor data using a nonlinear state estimation technique, which includes using nonlinear measurement models to update the navigation solution, providing a more accurate and robust navigation solution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional IMU/GNSS integration is used for navigation, then the system provides absolute position estimation that does not drift over time, but positioning accuracy deteriorates in dense urban areas where GNSS signals are blocked or affected by multipath
Solution Approach 1:
The patent combines IMU, GNSS, and visual odometry sensors into an integrated navigation system. The sensor fusion algorithm merges data from all three sources to produce a unified navigation solution that leverages the complementary strengths of each sensor type, achieving both reliability and accuracy in urban environments.
Solution Approach 2:
The navigation system is designed to universally handle multiple operating conditions by switching between and combining different sensor sources. The system can operate effectively whether GNSS signals are available or blocked, adapting to various urban environments through multi-functional sensor utilization.
2Measurement precision
If standalone IMU-based systems are used for pose estimation, then accurate relative pose estimation is achieved over short time periods, but errors accumulate exponentially with time
Solution Approach 1:
The system implements feedback mechanisms where GNSS provides periodic absolute position corrections to reset IMU drift, and visual odometry provides continuous feedback to constrain IMU error growth. This multi-loop feedback structure prevents error accumulation by regularly correcting the integrated navigation solution.
Solution Approach 2:
The navigation solution uses a composite estimation approach, combining results from IMU, GNSS, and visual odometry through a unified state estimation filter. This composite solution leverages the short-term accuracy of IMU while using the other sensors to prevent long-term error accumulation.
3Reliability
If GNSS signals are used for absolute positioning, then position estimation does not drift over time, but signals are completely blocked or affected by severe multipath in dense urban areas
Solution Approach 1:
Visual odometry acts as an intermediary sensor that provides positioning information when GNSS signals are blocked. The system uses visual features from the camera to estimate motion and position, serving as a mediator that bridges the gap when the primary GNSS positioning method becomes unavailable.
Solution Approach 2:
The system converts the harmful effect of GNSS signal blockage into an opportunity to utilize visual odometry. By designing the system to automatically switch to and integrate visual sensing when GNSS is unavailable, the harm of signal blockage becomes a trigger for activating alternative positioning methods that work specifically in those conditions.
4Measurement precision
If additional sensors are integrated to improve navigation accuracy in urban areas, then positioning reliability is enhanced, but device complexity increases
Solution Approach 1:
The navigation system is segmented into modular functional components: IMU processing module, GNSS processing module, visual odometry processing module, and sensor fusion module. Each module independently processes its respective sensor data and outputs standardized results to the fusion algorithm, reducing integration complexity through functional segmentation.
Data Source
AI summary
An integrated navigation solution is provided for a device within a moving platform. Motion sensor data from a sensor assembly of the device is obtained, optical samples from at least one optical sensor for the platform are obtained and map information for an environment encompassing the platform is obtained. Correspondingly, an integrated navigation solution is generated based at least in part on the obtained motion sensor data using a nonlinear state estimation technique, wherein the nonlinear state estimation technique uses a nonlinear measurement model for optical sensor data. Generating the integrated navigation solution includes using the sensor data with the nonlinear state estimation technique and integrating the optical sensor data directly by updating the nonlinear state estimation technique using the nonlinear measurement model and the map information. The integrated navigation solution is then provided.