Visual Odometry Fallback for GPS-Loss Autonomous Driving
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional autonomous vehicle navigation systems lack the ability to navigate routes if pre-loaded maps and/or GPS guidance systems fail, leading to situations where vehicles are blindly navigating without knowledge of their surroundings.
Innovation Solution
A novel computing system architecture that performs visual odometry using integrated cameras, allowing the autonomous vehicle to dynamically generate a map of its surroundings in real-time, even if GPS and pre-loaded maps are unavailable.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional autonomous vehicle navigation systems rely on pre-loaded maps and GPS guidance, then navigation accuracy is improved, but the system fails when GPS and pre-loaded maps are unavailable
Solution Approach 1:
The system changes the operational parameters of navigation by switching from GPS-coordinate-based navigation to visual-odometry-based navigation when GPS is unavailable. The computing system detects GPS signal loss and activates alternative visual navigation mode, changing how position and orientation are determined without relying on pre-loaded maps
Solution Approach 2:
The system prepares fallback navigation capabilities in advance by integrating multiple sensor systems (cameras, LiDAR, inertial measurement units) that can operate independently of GPS. This cushioning ensures continuous navigation capability even when primary GPS-based navigation fails
2Reliability
If the vehicle uses fallback visual odometry systems when GPS fails, then navigation reliability is improved, but computational complexity increases
Solution Approach 1:
The computing system is segmented into specialized modules: GPS signal monitoring module, visual odometry module, sensor fusion module, and navigation control module. Each module handles specific computational tasks, distributing the computational load and improving overall system efficiency when fallback navigation is activated
Solution Approach 2:
The system introduces an intermediary sensor fusion layer that integrates data from cameras, LiDAR, and inertial measurement units to bridge the gap between GPS failure and navigation continuity. This intermediary processing layer consolidates multiple data streams into unified position and orientation estimates, managing computational complexity
3Reliability
If the system continuously monitors GPS and activates fallback systems, then navigation reliability is improved, but energy consumption increases
Solution Approach 1:
The system implements periodic GPS signal monitoring at predetermined intervals rather than continuous monitoring. When GPS signal is detected as lost for a threshold period, the fallback visual odometry system is activated. This periodic check approach maintains navigation reliability while reducing unnecessary energy consumption during normal GPS-operational conditions
Data Source
AI summary
A system and method for performing visual odometry is disclosed. In aspects, the system implements methods to generate an image pyramid based on an input image received. A refined pose prior information representing a location and orientation of the autonomous vehicle can be generated based on one or more images of the image pyramid. One or more seed points can be selected from the one or more images of the image pyramid. One or more refined seed representing the one or more seed points with added depth values can be generated. One or more scene points can be generated based on the one or more refined seed points. A point cloud can be generated based on the one or more scene points.