Tele-Driving AI Data Prioritization for Low-Latency Remote Intervention
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous driving systems face challenges in smooth disengagement from autonomous mode to human control, particularly due to latency in sensory information relay and potential instability in wireless communication networks, which can lead to unreliable remote intervention and safety concerns.
Innovation Solution
A vehicular AI system that dynamically assesses communication channel properties and adapts data transmission by encoding, modifying, and prioritizing sensory data for efficient wireless communication, utilizing distributed AI resources and multiple communication links to ensure reliable remote operation and quick intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If sensory information is relayed remotely to human operators, then human intervention capability is improved, but latency increases and reliability deteriorates
Solution Approach 1:
The system segments control into multiple levels: autonomous driving unit handles normal operation, tele-driving unit handles remote intervention, and local human operator provides immediate response. This segmentation allows each component to operate optimally within its domain, improving overall system reliability while maintaining human intervention capability.
Solution Approach 2:
The tele-driving unit acts as an intermediary between the autonomous driving unit and remote operators. It receives sensory information locally, processes it through the tele-operating interface, and generates commands without requiring constant remote communication, thereby reducing latency and improving reliability.
2Ease of operation
If sensory information is transmitted wirelessly to remote operators, then remote monitoring capability is improved, but communication instability increases
Solution Approach 1:
The system separates critical control functions from non-critical monitoring functions. The tele-driving unit handles time-sensitive control commands locally, while wireless communication is used only for non-critical data transmission, reducing the impact of communication instability on overall system reliability.
Solution Approach 2:
The tele-driving unit autonomously processes sensory information and generates control commands without requiring continuous wireless communication with remote operators. This self-service capability ensures stable operation even when wireless communication is unstable or unavailable.
3Speed
If all sensory data is processed locally in the vehicle, then response speed is improved, but computational load increases
Solution Approach 1:
The system extracts only the essential processing functions needed for immediate response and keeps them in the vehicle's tele-driving unit. Non-critical data processing and storage are outsourced to remote servers, reducing local computational complexity while maintaining fast response times for critical decisions.
Solution Approach 2:
The system distributes computational tasks across multiple dimensions: local processing for immediate response, edge computing for intermediate processing, and cloud computing for comprehensive analysis. This multi-dimensional architecture balances response speed and computational complexity.
4Reliability
If multiple communication links are used for data transmission, then reliability is improved, but system complexity increases
Solution Approach 1:
The system merges multiple communication links at different hierarchical levels: short-range wireless for local tele-driving unit communication, cellular networks for remote operator connection, and wired infrastructure for backbone connectivity. This merged multi-layer approach improves reliability while managing complexity through clear separation of functions at each layer.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Device, system, and method of autonomous driving and tele-operated vehicles. A vehicular Artificial Intelligence (AI) unit, is configured: to receive inputs from a plurality of vehicular sensors of a vehicle; to locally process within the vehicle at least a first portion of the inputs; to wirelessly transmit via a vehicular wireless transmitter at least a second portion of the inputs to a remote tele-driving processor located externally to the vehicle; to wirelessly receive via a vehicular wireless receiver from the remote tele-driving processor, a remotely-computed processing result that is received from a remote Artificial Intelligence (AI) unit; and to implement a vehicular operating command based on the remotely-computed processing result, via an autonomous driving unit of the vehicle or via a tele-driving unit of the vehicle.