Teleoperations Guidance for Driverless Vehicle Trajectory Revisions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in navigating unpredictable events, such as construction zones or dynamic objects, where they may require guidance to safely maneuver or change their route, and existing systems lack effective collaboration between vehicles and remote operations for real-time decision-making.
Innovation Solution
A system that enables communication between driverless vehicles and a teleoperations system, allowing vehicles to request guidance and receive altered virtual boundaries for trajectory adjustments, enabling them to autonomously navigate through or around events by selecting revised trajectories based on confidence levels and sensor data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If autonomous vehicles operate independently without remote teleoperations system, then device complexity is reduced, but reliability deteriorates when encountering unpredictable events
Solution Approach 1:
A teleoperations system acts as an intermediary between autonomous vehicles and human operators. The system receives sensor data from vehicles, processes it through multiple modules (anomaly detection, semantic scene classification, trajectory generation), and provides guidance to vehicles when anomalies are detected, thereby improving reliability without requiring full human control
Solution Approach 2:
The teleoperations system is divided into independent functional modules: sensor data receiver, anomaly detector, semantic scene classifier, trajectory generator, and guidance provider. Each module handles specific aspects of the navigation problem, allowing the system to maintain reliability while managing complexity through modular architecture
2Reliability
If autonomous vehicles request guidance from teleoperations system for every event, then navigation reliability improves, but loss of time increases due to communication delays
Solution Approach 1:
The anomaly detection module selectively triggers teleoperations guidance only when specific anomalies are detected (e.g., construction zones, dynamic objects, traffic signals) rather than for every navigation event. This partial action approach maintains reliability for critical situations while avoiding unnecessary communication delays for routine scenarios
Solution Approach 2:
The semantic scene classification module pre-processes sensor data to identify and categorize potential anomalies before full trajectory planning occurs. By preliminarily classifying scenes and detecting anomalies early, the system prepares guidance information in advance, reducing actual response time when vehicles need assistance
3Adaptability or versatility
If teleoperations system provides detailed guidance for all scenarios, then adaptability improves, but device complexity increases
Solution Approach 1:
The trajectory generation module provides different types of guidance tailored to specific anomaly types: semantic scene classification identifies construction zones, dynamic objects, and traffic signals, with each triggering customized trajectory adjustments. This local quality approach ensures adaptability for diverse events while keeping control logic localized to specific anomaly handlers rather than requiring a single complex control system
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for operating a driverless vehicle may include receiving, at the driverless vehicle, sensor signals related to operation of the driverless vehicle, and road network data from a road network data store. The method may also include determining a driving corridor within which the driverless vehicle travels according to a trajectory, and causing the driverless vehicle to traverse a road network autonomously according to a path from a first geographic location to a second geographic location. The method may also include determining that an event associated with the path has occurred, and sending communication signals to a teleoperations system including a request for guidance and one or more of sensor data and the road network data. The method may include receiving, at the driverless vehicle, teleoperations signals from the teleoperations system, such that the vehicle controller determines a revised trajectory based at least in part on the teleoperations signals.