Autonomous Vehicle World Model Correction for Changing Road Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional autonomous vehicles rely on static, pre-processed map data that may become outdated due to changes in road features over time, leading to navigation inaccuracies and safety issues.
Innovation Solution
Generating and correcting a world model for autonomous vehicles using real-time or near real-time sensor data, incorporating both semantic and geometric corrections based on sensor data and static map data, to ensure accurate navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If static, pre-processed map data is used for navigation, then the system complexity is reduced and ease of operation is improved, but the reliability and accuracy of navigation deteriorate when road features change over time
Solution Approach 1:
The patent transforms the static map data system into a dynamic one by continuously updating the world model with real-time sensor data. The system dynamically adjusts the navigation model by comparing sensor-derived geometric features (lane markings, road edges, intersections) with stored map data, and updates the world model when discrepancies are detected, ensuring navigation accuracy adapts to changing road conditions while maintaining system usability
2Reliability
If real-time sensor data is continuously processed to update the world model, then the navigation accuracy and reliability are improved, but the computational complexity and energy consumption increase
Solution Approach 1:
The patent implements partial action by selectively updating only those portions of the world model that correspond to detected geometric features in the current sensor data. Rather than processing and updating the entire map continuously, the system performs corrections only for specific road segments where features are detected and compared with the stored world model, reducing computational load and energy consumption while maintaining navigation accuracy
Solution Approach 2:
The system employs feedback mechanisms by continuously comparing sensor-derived geometric features with the stored world model, detecting discrepancies, and using this information to selectively update the world model. This feedback loop ensures that energy is consumed only when and where updates are necessary, optimizing the balance between navigation accuracy and energy usage
Data Source
AI summary
Systems and methods of generating and updating a world model for autonomous vehicle navigation are disclosed. An autonomous vehicle system can receive sensor data from a plurality of sensors of an autonomous vehicle, where the sensor data is captured during operation of the autonomous vehicle; access a world model generated based at least on map information corresponding to a location of the operation of the autonomous vehicle; determine at least one semantic correction for the world model based on the sensor data; determine at least one geometric correction for the world model based on the sensor data and the map information; and generate an updated world model based on the at least one semantic correction and the at least one geometric correction.