Unmanned Vehicle Path Confidence for Adaptive Remote Assistance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles often struggle to safely navigate unfamiliar environments and complete tasks without human assistance, as their control systems may lack the ability to confidently identify viable navigation paths and avoid obstacles, leading to potential issues such as navigating through off-limits areas.
Innovation Solution
A machine learning model, such as an artificial neural network, is trained to determine navigation paths by initially receiving remote human control input, gradually gaining confidence to operate independently, and falling back on human assistance when needed, with the ability to adjust its confidence level based on terrain and available prior navigation paths.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the vehicle operates in autonomous mode without remote assistance, then productivity is improved, but reliability deteriorates due to insufficient confidence in navigating unfamiliar environments
Solution Approach 1:
A remote operator serves as an intermediary between the autonomous vehicle and the environment. The vehicle operates autonomously when confidence is high, but can request remote assistance when encountering unfamiliar situations. The remote operator provides guidance or takes control when needed, enabling the system to achieve both high productivity (through autonomous operation) and high reliability (through human oversight when necessary).
Solution Approach 2:
The navigation mode dynamically transitions between autonomous and remote-assisted based on confidence level. The system continuously evaluates its confidence in the current navigation path and adjusts the level of remote assistance accordingly. This dynamic adjustment allows the vehicle to maximize productivity while maintaining reliability by seeking help only when necessary.
2Reliability
If the vehicle requests remote assistance frequently, then reliability is improved, but loss of time increases due to communication delays and human intervention
Solution Approach 1:
The system changes the parameter of remote assistance frequency based on confidence level. When confidence is high, remote assistance is minimized or eliminated. When confidence drops below a threshold, remote assistance is requested. This parameter adjustment ensures reliability is maintained only when necessary, minimizing time loss while preventing navigation errors.
Solution Approach 2:
The autonomous vehicle serves itself by making navigation decisions independently when confident. It only requests remote assistance when it cannot confidently determine the correct path. This self-service approach minimizes reliance on remote operators, reducing communication delays and time loss while maintaining navigation accuracy.
3Adaptability or versatility
If the vehicle operates without learning from prior paths, then device complexity is reduced, but adaptability deteriorates in unfamiliar environments
Solution Approach 1:
The system performs preliminary actions by collecting and storing navigation paths from previous autonomous operations and remote-assisted operations. These prior paths serve as a knowledge base that the vehicle queries when encountering similar environments. This preliminary data collection enables the vehicle to adapt to new environments by leveraging experience from previous tasks, improving adaptability without requiring complex real-time learning algorithms.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Examples implementations relate to determining path confidence for a vehicle. An example method includes receiving a request for a vehicle to navigate a target location. The method further includes determining a navigation path for the vehicle to traverse a first segment of the target location based on a plurality of prior navigation paths previously determined for traversal of segments similar to the first segment of the target location. The method also includes determining a confidence level associated with the navigation path. Based on the determined confidence level, the method additionally includes selecting a navigation mode for the vehicle from a plurality of navigation modes corresponding to a plurality of levels of remote assistance. The method further includes causing the vehicle to traverse the first segment of the target location using a level of remote assistance corresponding to the selected navigation mode for the vehicle.