V2X Sensor Sharing for Blindspot-Focused 3D Perception
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for sharing sensor data between vehicles and remote systems face inefficiencies due to blindspots caused by occlusions and high bandwidth usage, which limits effective environmental perception for autonomous driving and driver assistive systems.
Innovation Solution
A system that uses a 3D sensor and controller to generate an initial 3D point cloud, classify cells in an occupancy grid map as drivable, undrivable, or blindspot cells, and selectively requests data from remote systems to fill blindspots, merging the received data with the initial point cloud to enhance environmental perception.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensor data is shared between vehicles and remote systems, then environmental perception accuracy is improved, but network bandwidth consumption increases
Solution Approach 1:
The patent segments the environment into an occupancy grid map with discrete cells, where only blindspot cells are identified and targeted for data requests. This segmentation allows selective data sharing rather than transmitting entire sensor datasets, reducing network bandwidth consumption while maintaining perception accuracy in critical areas.
Solution Approach 2:
The patent applies local quality by focusing data sharing efforts specifically on blindspot cells where environmental perception is degraded. Instead of uniformly sharing data across all areas, the system selectively requests and processes data only for locations where occlusions exist, optimizing both perception accuracy and network efficiency.
2Loss of information
If blindspots are filled by requesting data from remote systems, then environmental awareness is improved, but communication overhead increases
Solution Approach 1:
The patent performs preliminary action by pre-processing local sensor data to generate an occupancy grid map and identify blindspot cells before requesting remote data. This preparation allows the system to precisely target data requests to specific blindspot locations rather than requesting general environmental data, reducing communication overhead while filling information gaps.
Solution Approach 2:
The occupancy grid map serves as an intermediary structure that mediates between local sensor data and remote system data. It provides a standardized representation that enables efficient identification of blindspots and facilitates targeted data exchange with remote systems, reducing communication overhead while improving environmental awareness.
Data Source
AI summary
A system for sharing sensor data between a vehicle and a plurality of remote system generally includes a vehicle communication system, a three-dimensional (3D) sensor, and a controller. The controller is programmed to generate an initial 3D point cloud using the 3D sensor, generate an occupancy grid map based on the initial 3D point cloud, and select a producer remote system to provide data for a cell of the occupancy grid map classified as a first blindspot. The controller is further programmed to send a data request to the producer remote system using the vehicle communication system, receive data from the producer remote system including information about the cell of the occupancy grid map classified as the first blindspot, generate a merged 3D point cloud by merging the data received from the producer remote system with the initial 3D point cloud, and identify objects using the merged 3D point cloud.