Remote Assistance Command Toolbox for Autonomous Vehicle Edge Cases
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles often encounter difficulties in independently resolving issues such as object classification and navigation obstacles, necessitating remote assistance to ensure safe and efficient operation.
Innovation Solution
A computer-implemented method and system that allows autonomous vehicles to request remote assistance, with a computing system determining the necessary actions based on vehicle data and transmitting control signals to facilitate tasks like object classification and navigation through a customized user interface for remote operators.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If autonomous vehicles independently resolve issues using onboard systems, then device complexity and computational resource usage increase, but remote assistance capability and operational reliability improve
Solution Approach 1:
A remote assistance computing system acts as an intermediary between autonomous vehicles and human operators. The system receives requests from vehicles, determines appropriate assistance actions based on vehicle data, and transmits control signals back to vehicles. This mediator approach allows vehicles to maintain independent operation while providing access to remote human expertise when needed, resolving the contradiction between operational reliability and device complexity.
2Adaptability or versatility
If all remote assistance actions are displayed in the user interface, then operator versatility improves, but interface complexity and information overload increase
Solution Approach 1:
The remote assistance user interface displays different sets of actions based on the specific vehicle and situation. Rather than showing all possible actions uniformly, the system customizes the interface to show only relevant actions for each context. This localized customization maintains operator versatility while reducing interface complexity and preventing information overload.
Solution Approach 2:
The set of displayed remote assistance actions is dynamic and adapts based on vehicle data, request type, and operational context. The interface automatically adjusts which actions are available and visible, transitioning between different action sets as conditions change. This dynamic adaptation maintains versatility while managing interface complexity through context-aware filtering.
3Reliability
If autonomous vehicles request remote assistance frequently, then operational safety improves, but communication overhead and response time increase
Solution Approach 1:
The system prepares and pre-loads relevant vehicle data and potential assistance actions before they are needed. When a vehicle requests assistance, the computing system already has contextual information ready, enabling faster response. This preliminary preparation reduces communication overhead and response time while maintaining operational safety through proactive data gathering and action planning.
Data Source
AI summary
The present disclosure is directed to a system for generating customized command toolboxes for remote operators in a service system that includes autonomous vehicles. The system receives a request for remote assistance from an autonomous vehicle. The system determines, from a local storage location, vehicle data associated with the autonomous vehicle. The system selects a subset of remote assistance actions from a predetermined set of remote assistance actions. The system displays, in a remote assistance user interface, one or more user interface elements indicative of the subset of remote assistance actions. The system determines one or more remote assistance actions from the subset of remote assistance actions based at least in part on a user input associated with the one or more user interface elements. The system transmits one or more control signals associated with the one or more remote assistance actions.