Remote Vehicle Path Selection Using Pre-Generated Candidate Routes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles may become stuck and unable to navigate autonomously, making it challenging for remote vehicle assistance systems to quickly determine and provide acceptable alternate paths.
Innovation Solution
A remote vehicle assistance system that includes a dynamic graphical user interface for human operators to create, visualize, and manipulate candidate paths optimized based on costs and constraints, allowing for real-time selection and implementation of navigable paths by the vehicle.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If remote vehicle assistance systems use traditional path determination methods, then path selection can be made, but the time required to generate and implement alternate paths is excessive
Solution Approach 1:
The system pre-generates multiple candidate paths using path generation modules before they are needed. When the autonomous vehicle becomes stuck, these pre-computed paths are already available for immediate selection and implementation, eliminating the time delay associated with generating paths in real-time during the assistance operation.
Solution Approach 2:
The path determination process is divided into separate functional modules: path generation modules that create candidate paths, a presentation module that displays options to the remote operator, and a selection module that implements the chosen path. This segmentation allows each module to operate independently and efficiently, with pre-generated paths ready for rapid selection.
2Measurement precision
If remote vehicle assistance systems provide multiple candidate paths, then path selection accuracy improves, but system complexity increases
Solution Approach 1:
A presentation module acts as an intermediary between the path generation modules and the remote operator. This module receives multiple candidate paths from the generation modules, processes and displays them in a user-friendly format, and transmits the operator's selection to the implementation module. This intermediary simplifies the interface complexity while maintaining the ability to handle multiple paths for accurate selection.
3Reliability
If the system generates multiple candidate paths with optimization, then path quality improves, but computational requirements increase
Solution Approach 1:
Multiple candidate paths are generated and optimized in advance using computational resources before they are needed. The path generation modules perform the energy-intensive optimization calculations beforehand, so when the remote operator needs a path, the computationally expensive work is already complete and the optimized paths are ready for immediate use.
Data Source
AI summary
Provided are methods for drawing paths for remote vehicle assistance. The methods can include receiving a goal location responsive to an input provided via a graphical user interface and generating a plurality of candidate paths based on the goal location and a starting location of a vehicle. Each candidate path can represent a corresponding trajectory from the starting location to the goal location. The plurality of candidate paths can be provided via the graphical user interface as a representation of at least one corresponding cost and at least one corresponding constraint associated with each corresponding candidate path. A path selection can be received via the graphical user interface and data associated with the selected path can be provided to the vehicle to implement the corresponding trajectory. Systems and computer program products are also provided.