Autonomous Fleet Route Timing to Avoid Teleoperator Overlap
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Efficient and safe control of a fleet of autonomous/remotely operated vehicles is challenging due to the need for balancing the demand for human teleoperators and potential overlap in assistance or driving requirements, which can lead to inefficiencies and safety risks.
Innovation Solution
Assign pre-determined routes with road segments that require teleoperator assistance, determine time overlaps, and adjust the driving mode by inserting delays or speedups in autonomous segments to prevent simultaneous assistance demands, allowing for efficient utilization of teleoperators.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a human observer continuously monitors the autonomous vehicle, then the safety and reliability of the autonomous vehicle is improved, but the purpose of autonomous vehicle not needing a human driver is defeated
Solution Approach 1:
A teleoperator is introduced as an intermediary between the autonomous vehicle system and human operators. The teleoperator remotely monitors multiple vehicles and provides assistance only when needed, rather than continuously monitoring each vehicle. This mediator approach maintains high automation levels while providing safety backup through remote human intervention capability.
Solution Approach 2:
The teleoperator system is designed to handle multiple autonomous vehicles simultaneously through a single remote operation station. Instead of dedicating one human observer per vehicle, a universal teleoperator can assist multiple vehicles across different locations, maintaining safety while reducing the number of human operators needed and preserving autonomous operation.
2Reliability
If too many human observers or teleoperators are assigned to handle the fleet, then the risk of vehicles being stranded is reduced, but the system efficiency is reduced
Solution Approach 1:
The system implements partial monitoring and assistance where teleoperators only intervene when confidence levels drop below thresholds or specific situations arise. Instead of continuous full monitoring of all vehicles, the system applies human oversight selectively and partially, reducing the number of teleoperators needed while maintaining adequate coverage to prevent vehicles from being stranded.
Solution Approach 2:
The allocation of teleoperator resources is dynamic rather than static. The system adjusts the level of teleoperator involvement based on real-time conditions, vehicle confidence levels, and current fleet status. This dynamic approach allows the system to scale human intervention up or down as needed, optimizing efficiency while ensuring vehicles are not left without assistance when required.
3Productivity
If too few human observers or teleoperators are assigned to handle the fleet, then the system efficiency is improved, but the risk of vehicles being stranded increases
Solution Approach 1:
The system performs preliminary assessment of vehicle conditions and predicts when teleoperator assistance may be needed based on confidence levels and situational analysis. By anticipating potential issues before they become critical, the system can proactively allocate teleoperator resources to prevent vehicles from being stranded, rather than reacting after problems occur.
Solution Approach 2:
The autonomous vehicles continuously provide feedback to the teleoperator system about their operational status, sensor data quality, and confidence levels in their decision-making. This feedback loop allows the teleoperator system to monitor fleet conditions in real-time and dynamically reallocate limited teleoperator resources to vehicles that need assistance, preventing stranding while maintaining high efficiency with minimal human operators.
Data Source
Figure 1
Figure 2
Figure 3a~3b
AI summary
The present invention relates to a method for controlling a fleet (10) of at least two autonomous/remotely operated vehicles (12a-b), wherein the method comprises: assigning to each of said at least two vehicles (12a-b) a pre-determined route (36a-b) comprising a road segment (40a-b) which requires that a teleoperator (16) of a remote operation station (14) assists and/or drives the vehicle, determining a time overlap (42) of the said road segments (40a-b), and based on the determined time overlap, inserting a delay (44) or a speedup in a preceding autonomous drive road segment (38b) of the pre-determined route (36b) of at least one (12b) of said at least two autonomous/remotely operated vehicles so that said road segments (40a-b) no longer overlap in time.