Off-board Controller Data Segmentation for Real-time Machine Planning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing work site perception systems face challenges in real-time data utilization due to the large amount of data generated by machines, which exceeds the capacity for timely processing and transmission, especially in operating environments that limit data speed and quantity, hindering the sharing of perception information between machines.
Innovation Solution
A system that includes an off-board controller, position sensors, machine sensors, and perception sensors to determine the machine's pose and characteristics, generate point clouds, segment data, and transmit only relevant subsets off-board for updating electronic maps and generating machine movement plans, thereby reducing data transmission and improving real-time management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If all machine characteristics and perception data are transmitted off-board for real-time processing, then complete work site information is available, but data transmission time and network bandwidth are exceeded
Solution Approach 1:
The patent segments perception data into multiple data subsets based on machine characteristics and work site context. Only relevant subsets are transmitted off-board, while less critical data is processed locally or transmitted selectively. This segmentation reduces overall data transmission volume and time while preserving essential perception information for real-time decision-making.
Solution Approach 2:
The system extracts and transmits only the most critical perception data subsets off-board, rather than transmitting all raw perception data. By selecting and extracting only essential information based on machine characteristics and operational context, the system maintains real-time processing capability while minimizing data transmission time and bandwidth usage.
2Measurement precision
If all perception data is transmitted and processed, then complete work site mapping is achieved, but data processing capacity is exceeded
Solution Approach 1:
Perception data is segmented into multiple subsets with different levels of detail and importance. The system processes and transmits only necessary subsets for maintaining accurate work site maps, rather than processing all raw perception data. This selective processing maintains mapping accuracy while operating within data processing capacity limits.
Solution Approach 2:
The system applies different processing quality levels to different data subsets based on their importance and characteristics. Critical data subsets receive higher processing priority and more detailed analysis, while less critical data is processed with lower computational resources. This local quality approach maintains essential mapping accuracy while optimizing overall processing speed.
3Productivity
If comprehensive data is transmitted in real-time, then machine performance optimization is improved, but network bandwidth requirements are exceeded
Solution Approach 1:
Data for machine performance optimization is segmented into prioritized subsets, with only the most critical performance metrics and perception data transmitted in real-time. Less urgent data is transmitted with lower priority or processed locally. This segmentation enables machine performance optimization while keeping data transmission volume within network bandwidth constraints.
Solution Approach 2:
The system transmits a partial set of perception data that is sufficient for achieving machine performance optimization goals, rather than transmitting all available data. By identifying and transmitting only the essential data subsets needed for performance improvement, the system achieves effective optimization while minimizing data transmission volume.
Data Source
AI summary
A system for communicating between a machine and a remote system generates machine data and image data on-board the machine. The data is segmented into machine data subsets and image data subsets. At least some of the machine data subsets and only some of the image data subsets are transmitted off-board the machine to an off-board controller. The off-board controller updates an electronic map based upon the image data subsets, and generates a machine movement plan based upon the updated electronic map.