Vehicle-Pose Localization for Obstructed Urban Road Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In dense urban areas, autonomous vehicles (AVs) face challenges in localization due to sensor blockages, as onboard sensors like LiDAR and RADAR may be obstructed by other vehicles or road features, limiting their ability to extract sufficient information for navigation.
Innovation Solution
AVs employ a localization technique based on the pose of surrounding vehicles, estimating lane geometry and orientation from detected vehicle poses, which is then compared to map data to determine the vehicle's location and orientation, even in situations where primary sensors are obstructed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If onboard sensors (LiDAR, RADAR) are used for localization, then localization accuracy is improved under normal conditions, but sensor blockage by surrounding vehicles or road features causes localization failure in dense urban areas
Solution Approach 1:
The patent introduces surrounding vehicles as intermediary objects to mediate the localization process. Instead of directly using blocked sensors to determine own-vehicle pose, the system detects poses of surrounding vehicles and uses them as reference intermediaries to infer lane geometry and subsequently determine the own-vehicle's location and orientation through map data comparison.
Solution Approach 2:
The system creates a virtual copy of the road geometry by detecting and recording the poses of multiple surrounding vehicles. These vehicle pose data serve as copies or proxies for the actual lane geometry, allowing the system to reconstruct lane information indirectly when direct sensor observation is blocked.
2Device complexity
If traditional sensor-based localization is used, then the system is simple to implement, but it cannot extract sufficient information for navigation when sensors are blocked
Solution Approach 1:
The patent makes the localization system multi-functional by enabling it to operate in two distinct modes: direct sensor-based localization when sensors are unobstructed, and indirect vehicle-pose-based localization when sensors are blocked. This universality allows the same system to handle both normal and obstructed conditions without requiring separate dedicated systems.
Solution Approach 2:
The system uses surrounding vehicles, which are naturally present in the environment, to provide localization information. Instead of requiring additional specialized infrastructure or external services, the system leverages the vehicles already in the scene to serve the localization function, making the environment itself contribute to solving the localization problem.
Data Source
AI summary
Among other things, techniques are described for identifying sensor data from a sensor of a first vehicle that includes information related to a pose of at least two other vehicles on a road. The technique further includes determining a geometry of a portion of the road based at least in part on the information about the pose of the at least two other vehicles. The technique further includes comparing the geometry of the portion of the road with map data to identify a match between the portion of the road and a portion of the map data. The technique further includes determining a pose of the first vehicle relative to the map data based at least in part on the match.