Multi-Degree-of-Freedom Pose Estimation for GPS-Denied Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face navigation challenges in environments without reliable satellite-based positioning signals, such as GPS, leading to inadequate localization data and impaired autonomous operation.
Innovation Solution
An imaging system utilizing multiple electronic depth cameras and a deep neural network to determine a vehicle's multi-degree-of-freedom pose, allowing for navigation even in areas with weak or absent satellite signals by capturing images, processing them to determine the vehicle's position and sending navigation commands.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If satellite-based positioning signals are used for vehicle navigation, then positioning accuracy is improved, but the system fails in areas with weak or absent satellite signals
Solution Approach 1:
The patent introduces an imaging system with electronic depth cameras as an intermediary localization method. When satellite-based positioning is unavailable, the system captures images of the environment, processes them through deep neural networks to extract visual features, and determines vehicle position based on these visual landmarks, thereby providing a backup localization mechanism that operates independently of satellite signals
Solution Approach 2:
The system dynamically switches between different localization parameters and methods based on signal availability. It transitions from satellite-based coordinate parameters to visual feature-based position parameters processed through neural networks, changing the fundamental parameters used for determining vehicle location to maintain operation across different environmental conditions
2Reliability
If multiple electronic depth cameras and deep neural networks are deployed for localization, then navigation reliability in GPS-denied environments is improved, but device complexity increases
Solution Approach 1:
The imaging system serves multiple functions: it captures environment images for localization, provides depth information through electronic depth cameras, feeds data to deep neural networks for feature extraction, and generates navigation commands. This multi-functional approach consolidates what could be separate systems into a unified platform, managing complexity through functional integration rather than adding independent subsystems
Solution Approach 2:
The deep neural network is trained using images captured by the same imaging system, creating a self-training mechanism. The system uses its own captured data to improve its localization capabilities, reducing the need for external calibration equipment or manual intervention, thereby managing operational complexity through autonomous self-improvement
Data Source
AI summary
An imaging system and method of providing localization data to a vehicle using the imaging system is disclosed. The method may comprise: capturing, from an electronic depth camera, one or more images, wherein the one or more images include at least a portion of the vehicle; and using a deep neural network and the one or more images, determining a multi-degree of freedom (MDF) pose of the vehicle, wherein an optical axis of the electronic depth camera is oriented along a Z-axis according to a Cartesian coordinate system (comprising an X-axis, a Y-axis, and the Z-axis), wherein the Z-axis is plumb with respect to Earth.