Vehicle-Sensed Map Model Updates for Changing Road Geometry
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional road geometry modeling methods provide a single representation of road segments and objects, requiring resource-intensive data collection and human measurement, and are inadequate for dynamic environments where changes occur frequently, leading to inaccurate navigation for autonomous vehicles.
Innovation Solution
A method and apparatus for updating a map model by comparing received data from a vehicle to historical data to generate a modified internal model, incorporating changes in road geometry and objects, and enhancing navigation accuracy through confidence scoring and selective data transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional road geometry modeling methods are used to provide a single representation of road segments, then measurement precision may be maintained through human measurement and calculation, but resource consumption increases significantly and data collection becomes time-consuming
Solution Approach 1:
The patent uses sensor data from vehicles to create digital copies representations of road segments and objects, replacing traditional human measurement and calculation. Multiple vehicles collect data that is processed into map models, significantly reducing resource consumption while maintaining precision through automated sensing and processing systems.
Solution Approach 2:
The patent replaces manual human measurement and calculation with automated sensor-based systems. Vehicles equipped with sensors automatically capture road geometry and object data, eliminating the need for human operators to physically measure and calculate road features, thereby reducing resource consumption while maintaining measurement precision.
2Device complexity
If traditional single representation map models are used, then device complexity may be kept simple, but adaptability decreases in dynamic environments where road geometry and objects change frequently
Solution Approach 1:
The patent implements dynamic map models that are continuously updated based on sensor data from multiple vehicles. Instead of static single representations, the system maintains evolving models that adapt to changing road conditions, construction activities, and new objects, significantly improving adaptability while managing complexity through automated processing.
Solution Approach 2:
The patent divides the mapping system into multiple independent components: individual vehicle sensors, data transmission systems, processing servers, and model generation modules. This segmentation allows each component to operate independently and contributes to the overall adaptability without requiring complete system redesign when changes occur.
3Manufacturing precision
If resource-intensive traditional data collection methods are used, then manufacturing precision of map models may be maintained through careful human measurement, but productivity decreases due to time-consuming processes
Solution Approach 1:
The patent uses sensor data to create digital copies of road segments and objects, replacing traditional manual measurement processes. Multiple vehicles simultaneously collect data that is processed into precise map models, dramatically increasing productivity while maintaining precision through automated sensing and processing systems.
Solution Approach 2:
The patent implements continuous data collection and processing where vehicles constantly gather road information and the system continuously updates map models. This ongoing process eliminates the intermittent, time-consuming nature of traditional surveying methods, significantly improving productivity while maintaining precision through constant refinement of map data.
4Loss of information
If traditional map modeling approaches are used, then loss of information may be minimized through comprehensive human measurement, but reliability decreases in dynamic environments where road changes occur between measurement and navigation
Solution Approach 1:
The patent implements feedback loops where sensor data from vehicles continuously monitors road conditions and objects, comparing actual observations with existing map models. This feedback mechanism detects changes in real-time and triggers map model updates, ensuring navigation reliability by keeping map information current with the actual road environment.
Solution Approach 2:
The patent maintains continuous data collection and map model updating, ensuring that navigation information remains current. Instead of periodic updates, the system continuously refines map models based on ongoing sensor data, minimizing information loss and maintaining reliability in dynamic environments where road conditions change frequently.
Data Source
AI summary
Methods described herein relate to updating a map model. Methods may include receiving data from a vehicle pertaining to a location; comparing the received data pertaining to the location to historical data pertaining to the same location; and based at least in part upon a comparison of the received data pertaining to the location and the historical data pertaining to the same location, generating an internal model for the location. A corresponding apparatus and computer program product are also provided.