Remote Expert Mode for Autonomous Vehicle Obstacle Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face difficulties navigating around obstacles, particularly when lane markings interfere with their path, necessitating an expert mode for manual intervention to facilitate safe movement.
Innovation Solution
A system and method that allow manual instructions from a remote user to be implemented via a processor onboard the autonomous vehicle, utilizing a remote computer module and an autonomous vehicle computer module to determine assistance needs and facilitate movement around obstacles by generating alternate paths.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the autonomous vehicle relies solely on automatic navigation systems, then the extent of automation is high, but the ability to navigate around obstacles is insufficient when lane markings interfere with the path
Solution Approach 1:
The system dynamically switches between fully autonomous mode and expert mode with remote human assistance based on the complexity of the navigation situation. When obstacles require maneuvers that conflict with lane markings or require human judgment, the system transitions to expert mode, allowing dynamic adaptation of the automation level to match the situational demands
Solution Approach 2:
A remote user acts as an intermediary between the autonomous vehicle's automatic navigation system and the environment. When the autonomous system encounters obstacles it cannot handle, the remote user provides manual instructions that bridge the gap between automated operation and human decision-making capability
2Ease of operation
If the autonomous vehicle implements manual instructions from remote users, then the ease of operation improves for complex scenarios, but the device complexity increases due to additional communication and control systems
Solution Approach 1:
The complex decision-making capability is extracted from the vehicle itself and placed in a remote location. Instead of embedding sophisticated obstacle negotiation algorithms in the vehicle's onboard computer, the system extracts this function to human operators who can be accessed on-demand, reducing the computational complexity requirements of the vehicle's own systems
Solution Approach 2:
The remote user system serves multiple functions: it provides obstacle negotiation assistance, handles edge cases beyond the autonomous system's training, and can intervene in any situation where the vehicle requests help. This multi-functionality justifies the added communication infrastructure by consolidating various complex operations through a single remote assistance channel
Data Source
AI summary
Systems and method are provided for controlling an autonomous vehicle. In one embodiment, a method for controlling an autonomous vehicle comprises determining that assistance is required for continued movement of the autonomous vehicle; and facilitating movement of the autonomous vehicle via implementation, by a processor onboard the autonomous vehicle, of manual instructions provided from a remote user that is remote from the autonomous vehicle.