Crowdsourced Occlusion Map for Autonomous Vehicles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous and semi-autonomous vehicles face challenges in detecting occluded dynamic objects due to undetectable static objects and elevation changes, which can lead to unacceptably short reaction times and potential collisions, as their sensors' effective field of view and range are compromised by obstacles like buildings and road elevation.
Innovation Solution
A crowdsourcing system and method generate an occlusion map by collecting and processing perception data from multiple vehicles, using virtual detection lines and temporal classification data to identify non-occluded regions, allowing vehicles to avoid occluded areas and make informed routing decisions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensors are used to detect dynamic objects, then detection capability is improved, but effective field of view and range are reduced due to occlusions from static objects and elevation changes
Solution Approach 1:
The system performs preliminary mapping of occlusion zones using static object data and elevation information before dynamic object detection. By pre-identifying areas blocked by buildings, vegetation, and terrain features, the system compensates for sensor field of view limitations and maintains detection capability in occluded regions through predictive modeling and crowd-sourced historical data.
2Measurement precision
If LiDAR data or high-definition maps with 3-D information are used to improve occlusion detection, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The system creates simplified 2-D occlusion maps that replicate the essential occlusion information typically provided by complex 3-D LiDAR data and high-definition maps. By using crowd-sourced perception data from multiple vehicles to generate these maps, the system achieves accurate occlusion detection without requiring expensive specialized sensors or detailed three-dimensional mapping infrastructure.
3Reliability
If multiple sensor modalities and sensor mounting heights are used to overcome occlusions, then detection capability is improved, but device complexity and data processing requirements increase
Solution Approach 1:
The system merges perception data from multiple vehicles with different sensor configurations into a unified crowd-sourced occlusion map. By aggregating observations from diverse sensor modalities and mounting heights across the fleet, the system achieves comprehensive occlusion coverage that compensates for individual vehicle sensor limitations without requiring each vehicle to carry multiple specialized sensors.
Data Source
AI summary
A method for generating an occlusion map using crowdsourcing includes receiving occlusion data from a host vehicle. The occlusion data includes a plurality of virtual detection lines each extending from the sensor and intersecting an off-road area. The method further includes marking all non-occluded points that are located inside the off-road area and that are located within a predetermined distance from each plurality of virtual detection lines. The method further includes determining a non-occluded region inside the off-road area using the non-occluded points that have been previously marked inside the off-road area, wherein the non-occluded region is part of non-occlusion data. The method further includes generating an occlusion map using the non-occlusion data. The method further includes transmitting the occlusion map to the host vehicle. The method can also determine an occlusion caused by an elevation peak.