Vision-Based 3D Path Determination Without GPS Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles and robots often lack reliable navigation methods in environments where GPS data is unavailable, necessitating the ability to analyze surroundings and predict object presence without relying on location data.
Innovation Solution
Implementing AI models, such as occupancy and surface networks, to analyze images captured by cameras to determine voxel occupancy and generate 3D models, allowing self-localization and path planning without GPS.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If location data (GPS) is used for navigation, then path determination is straightforward, but the system cannot operate in environments where GPS is unavailable (indoor navigation)
Solution Approach 1:
The patent introduces an AI model as an intermediary that translates image data into occupancy predictions and 3D spatial understanding. This mediator enables the system to navigate without direct GPS input by creating a semantic map of the environment from visual observations alone
Solution Approach 2:
The patent replaces the GPS-based mechanical localization system with an AI-based visual localization system. Instead of relying on satellite signals and coordinate systems, the system uses neural networks to interpret images and infer spatial relationships, enabling operation in GPS-denied environments
2Adaptability or versatility
If AI models are used to predict occupancy and generate 3D models from images, then GPS-free navigation is enabled, but computational complexity and processing time increase
Solution Approach 1:
The patent pre-trains AI models on large datasets of images and occupancy information before deployment. This preliminary training allows the models to make accurate predictions with fewer computational resources during actual navigation, reducing real-time processing complexity
Solution Approach 2:
The patent transforms 2D image data into 3D occupancy predictions and spatial models using neural networks. By adding the depth dimension through AI processing, the system creates comprehensive spatial understanding from limited visual input, improving navigation capability without proportionally increasing computational load
Data Source
AI summary
Disclosed herein are methods and systems for using artificial intelligence modeling techniques to generate a path for an ego. In an embodiment, a method comprises retrieving image data of a space around an ego, the image data captured by a camera of the ego; predicting by executing an artificial intelligence model, an occupancy attribute of a plurality of voxels corresponding to the space around the ego; generating a 3D model corresponding to the space around the ego and each voxel's occupancy attribute; upon receiving a destination, localizing, by the processor, the ego by identifying a current location of the ego using a key image feature within the image data corresponding to the 3D model without receiving a location of the ego from a location tracking sensor; and generating a path for the ego to travel from the current location to the destination.