Hybrid Drivable Path Mapping for Routeless AV Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in navigating due to the vast volume of data needed for traditional mapping, which can limit or adversely affect navigation, and the sheer quantity of data poses daunting challenges for storage and updates.
Innovation Solution
A system that generates map information using drive information from multiple vehicles to create target trajectories, aggregates road topography features, and combines these with crowdsourced trajectory data to provide hybrid trajectories for navigation, utilizing trained models and actuators for vehicle navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional mapping technology is used for autonomous vehicle navigation, then navigation accuracy can be maintained, but the volume of data required for storage and updates becomes excessively large
Solution Approach 1:
The patent extracts only the essential navigational elements from traditional comprehensive maps, creating a sparse map that contains only critical trajectory information, intersection data, and stand-on trajectory identifiers needed for autonomous navigation, thereby dramatically reducing data volume while preserving navigation accuracy
Solution Approach 2:
The patent creates simplified copies of traditional map data by generating target trajectories from road topography features and combining them with crowdsourced trajectory data to form hybrid trajectories, which are stored as compact representations in the sparse map structure
2Reliability
If comprehensive map data is stored and updated continuously, then navigation reliability is improved, but the complexity of data management and system operations increases
Solution Approach 1:
The patent segments the navigation data structure into distinct components: road topography features, target trajectories derived from topography, crowdsourced trajectories from multiple vehicles, and their combination as hybrid trajectories. This segmentation allows each component to be managed independently with simplified update procedures
Solution Approach 2:
The system enables autonomous vehicles to contribute their own trajectory data to the sparse map through crowdsourcing, allowing the map to self-update and self-improve without requiring centralized manual updates, thereby reducing operational complexity while maintaining reliability
Data Source
AI summary
In one implementation, a system generates a map for use in navigating a host vehicle relative to a road segment. The system may receive drive information from each of a plurality of vehicles that traversed the road segment. The system may aggregate indicators representative of road topography features and generate a representation of road topography of the road segment based on the aggregated indicators; provide the representation of road topography of the road segment as input to at least one trained model configured to generate, in response to the provided input, an output including a target trajectory for at least a first portion of the road segment; aggregate the actual trajectory information included in the drive information received from the plurality of vehicles that traversed the road segment; generate a crowdsourced trajectory for at least a second portion of the road segment based on the aggregated actual trajectory information; combine the target trajectory and the crowdsourced trajectory to generate a hybrid trajectory associated with the road segment; store the hybrid trajectory in the map; and provide the map to at least one host vehicle navigation system.