Teleoperator Situational Awareness for Predictive Vehicle Assistance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Fully-autonomous vehicles without driving controls face challenges in navigating unpredictable scenarios, leading to potential delays or disruptions in traffic flow, as they may slow or stop unpredictably, causing irritation to passengers and other drivers.
Innovation Solution
A machine-learned situational awareness engine is trained to predict when teleoperator assistance is needed and to present relevant guidance to a remote operator, optimizing the duration of assistance requests and minimizing operator interactions by using vehicle data, sensor data, and historical interaction data to configure the presentation and options for real-time decision-making.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If fully-autonomous vehicles are equipped without driving controls, then automation level is improved, but the vehicle may slow or stop unpredictably causing irritation and traffic disruption
Solution Approach 1:
The system performs preliminary actions by predicting future teleoperation requests before they occur. The machine-learned model analyzes current vehicle state, sensor data, and historical patterns to anticipate when assistance will be needed, allowing the vehicle to prepare and reduce unnecessary stops or slowdowns while maintaining high automation.
Solution Approach 2:
The system implements feedback by continuously monitoring vehicle operation and comparing predicted teleoperation needs against actual outcomes. Historical teleoperation data is fed back into the model to refine predictions, enabling the vehicle to learn from past interactions and improve its automation decisions to minimize traffic disruption.
2Reliability
If teleoperator assistance is provided frequently, then safety is improved, but time loss and traffic impeded increases
Solution Approach 1:
The system performs preliminary action by predicting which teleoperation requests are truly necessary before the vehicle encounters uncertain scenarios. The machine-learned model filters out routine situations that don't require human assistance, allowing the vehicle to operate autonomously in predictable conditions while maintaining safety by seeking help only when genuinely needed.
Solution Approach 2:
The system applies partial action by providing teleoperator assistance only for the specific portion of operation that requires human judgment. Rather than constant human oversight, the model identifies precise moments when teleoperation is beneficial, optimizing the balance between safety and time efficiency.
3Loss of time
If machine-learned model predicts teleoperation requests accurately, then assistance time is reduced, but model complexity increases
Solution Approach 1:
The machine-learned model serves multiple functions simultaneously: it predicts teleoperation requests, classifies scenario types, identifies key sensor data patterns, and generates recommendations for optimal assistance timing. This multi-functionality reduces the need for separate systems while minimizing assistance time through integrated decision-making.
Solution Approach 2:
The system performs self-service by using its own historical teleoperation data and sensor information to train and refine its predictive model. The vehicle learns from its own operational patterns and teleoperation interactions, reducing the need for external configuration or complex manual tuning while improving prediction accuracy over time.
Data Source
AI summary
A teleoperator device may be configured to obtain a request for teleoperator assistance from a driverless vehicle and obtain teleoperator data in response to the request. The teleoperator device may also be configured to record at least some of the teleoperator input and/or guidance transmitted to the driverless vehicle based on the teleoperator input. Upon receiving a subsequent request, the teleoperator device may be configured to reproduce at least part of the former teleoperator input and/or to provide an option to activate guidance associated with the teleoperator input. The teleoperator device may also be configured to train a model and/or use a model to determine from vehicle data an option for presentation via a teleoperator interface and/or a presentation configuration of the teleoperator interface.