Embedded Map Instructions for Local Sensor Data Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Consumer vehicles capture vast amounts of detailed on-board road sensor data for assisted and autonomous driving, but most of this data is discarded due to volume and privacy concerns, limiting map learning and updates.
Innovation Solution
Embedding computer-executable instructions in digital maps allows on-board vehicles to process and analyze sensor data locally, transmitting only analyzed results, thus updating maps in real-time without large data transmission and addressing privacy issues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If consumer vehicles transmit all captured sensor data for map learning, then map accuracy improves, but data transmission volume and privacy concerns increase
Solution Approach 1:
The patent extracts only the essential analysis results from the captured sensor data rather than transmitting the complete raw data sets. The on-board apparatus processes the sensor data locally and transmits only the extracted findings to the map building apparatus, significantly reducing transmission volume while preserving map learning effectiveness
Solution Approach 2:
The patent introduces an on-board data processing apparatus as an intermediary between the sensor data capture and map building processes. This intermediary performs local analysis and filtering of sensor data, transforming raw high-volume data into condensed, processed results that can be transmitted efficiently while maintaining data quality for map updates
2Measurement precision
If consumer vehicles store all captured sensor data, then map learning quality improves, but storage requirements and privacy risks increase
Solution Approach 1:
The system extracts only the necessary information from captured sensor data for map learning purposes. The on-board apparatus identifies and extracts relevant features and anomalies from the sensor data, storing only these extracted elements rather than the complete raw data sets, thereby reducing storage requirements while maintaining learning quality
Solution Approach 2:
The on-board vehicle apparatus performs self-service data processing by automatically analyzing and filtering sensor data locally. The system independently determines which data elements are valuable for map learning and processes them on-site, eliminating the need to store and transmit unnecessary data while ensuring high-quality map learning
3Measurement precision
If detailed sensor data is transmitted for processing, then map update accuracy improves, but transmission bandwidth requirements increase
Solution Approach 1:
The patent applies data extraction by removing unnecessary components from the sensor data before transmission. The on-board apparatus extracts only the critical findings and relevant features needed for map updates, significantly reducing the transmission data volume and bandwidth requirements while preserving the accuracy needed for effective map learning
4Productivity
If all captured sensor data is analyzed, then map learning effectiveness improves, but processing complexity and energy consumption increase
Solution Approach 1:
The system extracts only the essential elements from captured sensor data that are relevant for map learning. The on-board apparatus identifies and extracts key features, anomalies, and critical information while discarding redundant data, thereby maintaining map learning effectiveness while significantly reducing processing complexity and energy requirements
Solution Approach 2:
The patent applies partial action by analyzing only the portion of sensor data that is necessary for map learning rather than processing all captured data. The system selectively processes specific data elements based on their relevance to map updates, achieving effective map learning with reduced processing complexity and energy consumption
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
Methods, apparatuses, systems, and computer program products are provided. An example method comprises accessing a record for a particular traversable map element. The record comprises an executable instruction. The example method further comprises receiving sensor data from one or more sensors. The sensor data corresponds to the particular traversable map element. The method further comprises executing, by a processor, the executable instruction. Executing the executable instruction causes analysis of at least a portion of the sensor data.