Perceptive AV Navigation Using Asset Features Without GPS
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles (AVs) face challenges in navigating accurately without GPS, particularly in GPS-denied environments, where features of interest are underground, indoors, or under bridges, requiring more precise location accuracy than GPS can provide.
Innovation Solution
A navigation system and method for AVs that utilizes onboard generated information, such as image data, to determine positional pose and orientation in three-dimensional space, associating it with a frame of reference, and using environmental sensors like cameras and LIDAR to detect asset features for perceptive navigation, enabling motion control commands to be generated without relying on GPS.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS-based location determination is used, then the navigation system is simple and easy to operate, but the location accuracy is insufficient (approximately 2 meters) for civilian inspection applications requiring accuracy within 2 meters
Solution Approach 1:
The patent introduces an intermediary system consisting of visual markers and computer vision algorithms that mediate between the AV and the environment. These markers serve as intermediate reference points that the AV can detect and use for precise localization, transforming the direct GPS-to-position problem into a multi-step process involving marker detection, feature extraction, and pose estimation, thereby achieving sub-2-meter accuracy without direct GPS dependency
Solution Approach 2:
The patent replaces the GPS satellite-based electromagnetic positioning system with an onboard computer vision system using cameras and visual markers. This substitution eliminates dependency on external satellite infrastructure and enables precise localization in GPS-denied environments by using optical field-based marker detection and image processing algorithms
2Adaptability or versatility
If GPS is used for navigation, then the system works in open environments, but it fails in GPS-denied environments such as underground, indoors, or under bridges
Solution Approach 1:
The patent creates a universal navigation system that can operate across multiple environments (GPS-available, GPS-denied, indoor, outdoor, underground) by integrating multiple localization approaches. The system uses visual markers as a universal reference system that functions independently of GPS availability, making the navigation reliable across diverse and changing environmental conditions
Solution Approach 2:
The patent implements preliminary action by pre-deploying visual markers in the environment before the AV arrives. These markers are installed in advance to create a known reference framework that the AV can detect and use for localization, eliminating the need for real-time GPS signals and enabling immediate navigation upon entering GPS-denied areas
3Measurement precision
If the AV operates without precise location determination, then the system is simpler to operate, but it cannot achieve accuracy equivalent to human inspectors for civilian inspection applications
Solution Approach 1:
The patent implements self-service by enabling the AV to autonomously determine its own position and orientation through onboard cameras and computer vision algorithms. The system automatically detects visual markers, extracts features, calculates pose information, and updates its localization without human intervention, achieving high positioning precision while maintaining ease of operation through automated processes
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables AVs to navigate with accuracy equivalent to human inspectors, achieving location precision within 2 meters, even in GPS-denied environments, by using onboard sensors and machine perception to identify and respond to asset features.
Implementation Method 1
using onboard sensors and machine perception to identify and respond to asset features
Data Source
AI summary
A navigation system and a navigation method for an autonomous vehicle (AV) includes a system controller and an environmental sensor. The system controller determines an AV positional pose, which identifies the location of the AV, which further includes an AV position and an AV orientation in three-dimensional space. Additionally, the system controller determines a frame of reference that is associated with the AV positional pose. The frame of reference includes a coordinate system for the AV position and the AV orientation. A localization module determines the AV positional pose and the corresponding frame of reference. The localization module is associated with the system controller. The system controller then identifies at least one asset feature frame (AFF), which associates the AV positional pose with the feature in the AV environment. The system controller generates a motion control command based the AV positional pose and the AFF.