Collaborative 3-D Map for Autonomous Vehicles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Computer-assisted and autonomous driving vehicles face challenges in constructing a comprehensive 3-D view of their environment due to occlusions by other vehicles and objects, which can lead to incomplete situational awareness and increased risk of accidents, especially when different vehicles use varying techniques for object detection and classification.
Innovation Solution
A collaborative 3-D mapping system that allows vehicles to share and integrate sensor data from neighboring vehicles, using a consensus-based approach to classify objects and leverage diverse neural networks to enhance accuracy and resilience against adversarial attacks, thereby creating a comprehensive and accurate 3-D map of the environment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single vehicle constructs its own 3-D map using its own sensors, then the system complexity is low, but the measurement precision and completeness of the environment map deteriorates due to occlusions
Solution Approach 1:
The patent merges 3-D map data from multiple CA/AD vehicles to create a collaborative environment map. Each vehicle contributes its sensor data and detected objects, combining individual partial views into a comprehensive collective map that overcomes occlusion limitations and improves measurement precision without requiring each vehicle to have complex individual systems
Solution Approach 2:
The patent introduces a communication system as an intermediary that facilitates data exchange between vehicles. This mediator enables the sharing of 3-D map information, object detections, and sensor data, allowing vehicles to collaborate on environment mapping without direct complex inter-vehicle coordination
2Adaptability or versatility
If vehicles use diverse object detection techniques, then the adaptability improves, but the reliability deteriorates due to inconsistencies in object classification
Solution Approach 1:
The patent implements a feedback mechanism where vehicles share their object detections and classifications with the collaborative map system. The system processes this feedback from multiple sources, compares classifications, and resolves inconsistencies through consensus algorithms, thereby maintaining reliability while preserving the benefits of diverse detection techniques
Solution Approach 2:
The patent creates a composite classification approach by combining results from multiple diverse detection techniques. Rather than relying on a single method, the system integrates classifications from various vehicles using different detection algorithms, creating a more robust and reliable consensus classification that leverages the strengths of each approach
3Loss of information
If vehicles share comprehensive sensor data, then the quantity of information increases, but the loss of time in data processing increases
Solution Approach 1:
The patent extracts and shares only the essential and relevant portions of sensor data needed for collaborative mapping, rather than transmitting complete raw sensor streams. By extracting key information such as detected objects, their classifications, positions, and critical environmental features, the system maintains data completeness while significantly reducing processing time and communication overhead
Data Source
AI summary
Disclosures herein may be directed to a method, technique, or apparatus directed to a computer-assisted or autonomous driving (CA/AD) vehicle that includes a system controller, disposed in a first CA/AD vehicle, to manage a collaborative three-dimensional (3-D) map of an environment around the first CA/AD vehicle, wherein the system controller is to receive, from another CA/AD vehicle proximate to the first CA/AD vehicle, an indication of at least a portion of another 3-D map of another environment around both the first CA/AD vehicle and the another CA/AD vehicle and incorporate the at least the portion of the 3-D map proximate to the first CA/AD vehicle and the another CA/AD vehicle into the 3-D map of the environment of the first CA/AD vehicle managed by the system controller.