Autonomous Vehicle Map Change Detection for Real-Time Semantic Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face navigation issues due to outdated maps, which are not updated quickly enough to reflect changing road conditions, leading to restricted areas and inefficiencies in route planning and execution.
Innovation Solution
A system that enables autonomous vehicles to update semantic labels of their maps in real-time using sensor data, allowing for detection and adaptation to changes in lane lines and other road features without the need for a special purpose mapping vehicle, thereby reducing downtime and workload on such vehicles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If maps are updated using special purpose mapping vehicles, then map data quality and resolution are improved, but update time and operational efficiency deteriorate
Solution Approach 1:
The patent uses sensor data from autonomous vehicles as a copy or approximation of the high-resolution map data that would be collected by special purpose mapping vehicles. Instead of deploying expensive specialized equipment, the system captures sufficient navigation information using sensors already present on regular autonomous vehicles, thereby obtaining updated map data without the time loss and cost associated with dedicated mapping expeditions.
Solution Approach 2:
Autonomous vehicles perform map updating as part of their normal operational routine. While conducting regular navigation tasks, these vehicles simultaneously collect and contribute map data, eliminating the need for separate mapping missions. The system transforms routine vehicle operations into a dual-purpose activity that both navigates and updates maps, thereby resolving the contradiction between update frequency and operational efficiency.
2Reliability
If maps are updated frequently using special purpose mapping vehicles, then map currency is improved, but workload and resource allocation deteriorate
Solution Approach 1:
The patent makes autonomous vehicles multi-functional by enabling them to perform both their primary navigation function and the secondary function of map updating. This eliminates the need for separate special purpose mapping vehicles and complex coordinated mapping operations. The same vehicle infrastructure, sensors, and operational framework serve dual purposes, thereby maintaining map currency while reducing overall system complexity and resource requirements.
Solution Approach 2:
Instead of using specialized mapping vehicles with dedicated equipment, the system uses regular autonomous vehicles with standard sensors to collect map data. This copying approach uses existing infrastructure for a new purpose, avoiding the complexity of deploying and managing specialized mapping operations while achieving the same goal of maintaining current map data.
3Measurement precision
If high-resolution map data is collected continuously, then navigation accuracy is improved, but data processing load and energy consumption deteriorate
Solution Approach 1:
The patent applies partial action by selectively processing only those sensor data elements that contain actual map changes or anomalies. Instead of continuously processing all sensor data at full resolution, the system identifies and processes only relevant portions—such as detected lane markings, road features, or inconsistencies with existing maps—thereby maintaining navigation accuracy while significantly reducing computational load and energy consumption compared to continuous full-data processing.
Data Source
AI summary
The present technology provides systems, methods, and devices that can update aspects of a map as an autonomous vehicle navigates a route, and therefore avoids the need for dispatching a special purpose mapping vehicle for these updates. As the autonomous vehicle navigates the route, data captured by at least one sensor of an autonomous vehicle can indicate an inconsistency between pre-mapped from a high-resolution sensor system describing a location on a map, and current data describing a new feature of the location. A type of the new feature can be classified in accordance with an analysis of a structure of the current data that has been clustered together based on a threshold spatial closeness, and semantic labels of the pre-mapped data from the high-resolution sensor system can be updated based on the new feature described by the clustered current data.