Remote Vehicle Control Intent Prediction for Lost-Link Jitter
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing remote operation systems for vehicles face challenges in handling communications jitter and intermittency, particularly in ensuring safe and efficient operation of dynamically-unstable vehicles when the communication link is lost or becomes stale.
Innovation Solution
The system predicts the remote operator's intent over long time scales by combining short-horizon predictions based on input device position data with real-time sensor data and safety objectives, using an explicit optimization-based algorithm to generate control signals for the vehicle.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the lost-link timeout threshold TLL is set too short, then the lost link protocol is triggered more frequently, but it may trigger the autonomous mode due to relatively benign jitter and packet drops that invariably occur over complicated networks
Solution Approach 1:
The system uses feedback from the communication link quality and vehicle state to intelligently adjust the timeout threshold TLL. By monitoring packet delivery patterns, latency, and other communication metrics, the system can distinguish between benign jitter and actual link loss, triggering the lost-link protocol only when truly necessary rather than reacting to normal network variations.
Solution Approach 2:
The system changes the timeout threshold parameter TLL dynamically based on communication conditions and vehicle context. Rather than using a static threshold, the system adapts this critical parameter in real-time to match the actual operational environment, extending it during stable conditions to avoid false triggers while maintaining sensitivity during degraded conditions.
2Device complexity
If conventional zero-order hold is used to handle stale input, then the system is simple to implement, but it is not appropriate for vehicles with unstable dynamics operating in fast-changing environments
Solution Approach 1:
Instead of simply holding the latest input value, the system performs preliminary computation of operator intent using prediction algorithms that model the operator's future commands. This pre-computed intent prediction provides a more accurate representation of what the operator would command next, giving unstable vehicles a head start in maintaining stability during communication delays.
Solution Approach 2:
The intent prediction algorithm acts as an intermediary between the operator's actual inputs and the vehicle's control system. Rather than directly transmitting raw control inputs or simply holding them, the prediction algorithm processes and transforms these inputs into predicted future commands, smoothing out the control signal for unstable vehicles while preserving the operator's intent.
Data Source
AI summary
The present invention provides methods and systems for remote operation of a vehicle with the capability to deal with communications jitter and intermittency. In particular, the methods and systems herein may safely predict a remote operator's intent (e.g., remote pilot) over long time scales, and up to the lost link timeout TLL.