Autonomous Vehicle Map Update via Landmark Visibility Ranges
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing map updating methods for autonomous vehicles are inadequate as they often result in outdated maps due to changes in the environment, such as roadworks, leading to unreliable self-localization when landmarks recorded in the map no longer exist or are concealed.
Innovation Solution
A method that involves detecting landmarks using sensors and transmitting a deletion hypothesis to a central computer, which verifies the hypothesis by gathering data from multiple vehicles, reducing unnecessary detection and transmission costs, and indicating visibility ranges to optimize data processing and reduce power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If continuous sensor detection and data transmission is performed to update maps, then map accuracy is improved, but energy consumption and detection costs increase
Solution Approach 1:
The system pre-calculates and stores visibility ranges for landmarks in the map before actual detection occurs. This preliminary preparation allows the vehicle to later determine whether sensor detection is necessary based on pre-computed visibility information, avoiding unnecessary energy-consuming sensor operations while maintaining map accuracy.
Solution Approach 2:
Instead of continuously performing full sensor detection and data transmission, the system performs partial detection only when necessary - specifically when the vehicle enters a landmark's visibility range. This selective approach reduces energy consumption and detection costs while still maintaining sufficient map accuracy for autonomous driving operations.
2Measurement precision
If sensor detection is performed for all landmarks, then detection accuracy is improved, but detection costs and processing time increase
Solution Approach 1:
Visibility ranges for all landmarks are pre-calculated and stored in the map data structure before the vehicle reaches them. This preliminary computation includes geometric relationships, occlusion information, and detection feasibility assessments, enabling efficient real-time decision-making about which landmarks require actual sensor detection.
Solution Approach 2:
The detection process is segmented into two phases: (1) pre-computation phase where visibility ranges are calculated offline and stored in the map, and (2) runtime phase where only specific landmarks within visibility ranges undergo actual sensor detection. This segmentation separates complex computational tasks from real-time operations, improving overall detection efficiency.
3Reliability
If map updates are performed frequently, then map currency is improved, but data transmission volume and processing load increase
Solution Approach 1:
The system extracts only the essential visibility range information from complete landmark data and stores it in the map. By taking out only the critical geometric and visibility parameters needed for detection decisions, the system reduces data transmission volume while maintaining the ability to perform accurate detection assessments and update maps efficiently.
Data Source
AI summary
A method for updating a map of the surrounding area wherein the map is used when controlling a transportation vehicle. The transportation vehicle attempts, when travelling along a carriageway, to detect a landmark recorded in the map of the surrounding area using sensors. In response to the transportation vehicle being unable to detect the landmark using its sensors, the transportation vehicle prepares a proposal for deletion which is sent to an external central computer. The proposal for deletion is examined in the external central computer and deletes the landmark from the map of the surrounding area in response to the proposal for deletion being verified. In the method, information regarding a visibility range for the landmark is entered in the map of the surrounding area specifying at what range the landmark is visible from a road permitting a more effective detection of landmarks.