Dynamic Route Mapping for Autonomous Vehicle Behavior Uncertainty
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional route planning for autonomous and semi-autonomous vehicles requires extensive and costly data collection, which is inefficient and impractical, especially when deploying vehicles to new geographic areas with different driver behaviors.
Innovation Solution
A system and method that generate dynamic maps and routes by approximating initial map information from a database, using sensors to collect and update data, and comparing it to similar datasets to reduce uncertainty in road-agent behavior, allowing for efficient data collection and route optimization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If arbitrary routes are driven to collect driving data in new geographic areas, then complete driver behavior data can be obtained, but the data collection process becomes extremely time-consuming and costly
Solution Approach 1:
The system performs preliminary actions by generating an initial map using feature datasets from similar geographic areas before actual data collection begins. This pre-population of the map with approximate data allows the vehicle to start with useful information rather than complete emptiness, significantly reducing the time needed to collect sufficient driver behavior data for reliable route planning
Solution Approach 2:
The system creates copies of existing feature datasets from similar geographic areas to populate the initial map. By copying and adapting data from analogous regions, the system obtains a head start in understanding driver behaviors without having to collect all data from scratch in the new geographic area, thereby reducing data collection time while maintaining data reliability
2Reliability
If arbitrary routes are driven to collect driving data, then comprehensive behavioral data can be gathered, but the cost of deployment increases significantly
Solution Approach 1:
The system performs preliminary actions by generating an initial map using feature datasets from similar geographic areas before actual data collection begins. This pre-population of the map with approximate data allows the vehicle to start with useful information rather than complete emptiness, significantly reducing the time needed to collect sufficient driver behavior data for reliable route planning
Solution Approach 2:
The system creates copies of existing feature datasets from similar geographic areas to populate the initial map. By copying and adapting data from analogous regions, the system obtains a head start in understanding driver behaviors without having to collect all data from scratch in the new geographic area, thereby reducing data collection time while maintaining data reliability
3Measurement precision
If the system waits to collect sufficient data before generating routes, then route planning accuracy improves, but deployment efficiency decreases
Solution Approach 1:
The system dynamically updates the map as data is collected, transitioning from a static initial map to an increasingly accurate dynamic map. This allows the system to generate routes at multiple stages - initially using approximate data for quick deployment, then continuously improving route planning accuracy as more data is collected, thereby balancing deployment efficiency with measurement precision
Solution Approach 2:
The system implements feedback by continuously comparing newly collected feature datasets with existing map data and updating the map accordingly. This feedback loop allows the system to maintain decent route planning accuracy from the start using the initial map, then progressively improve accuracy as feedback from actual data collection refines the map, thus maintaining both deployment efficiency and measurement precision
Data Source
AI summary
Systems and methods for generating efficient planning routes for vehicles, including autonomous and semi-autonomous vehicles are presented. A route planner may generate dynamic maps and routes that reduces the uncertainty of road-agent environmental and behavioral data in an efficient manner. Route planning may be accomplished using a statistical approach in which known data from one geographic or behavioral feature set may be used and relied upon by a vehicle in another geographical and behavioral context to estimate the environmental and behavioral data relevant to the vehicles current operation.