Tele-Operated Driving

Description

Tele-Operated Driving (TOD) is a service defined within 3GPP standards that facilitates the remote control of vehicles over cellular networks. It is a key use case for 5G and beyond, demanding stringent performance requirements, particularly ultra-low latency and high reliability, to ensure safe and responsive remote operation. The service involves a remote human operator or automated system receiving real-time sensor data (video, LiDAR, radar) from the vehicle and sending corresponding control commands (steering, acceleration, braking) back. The 3GPP system provides the communication framework, with specifications detailing the necessary quality of service (QoS) flows, positioning requirements for the remote operator and vehicle, and performance evaluation methodologies.

Architecturally, TOD leverages the 5G system's support for URLLC and network slicing. A dedicated network slice can be instantiated to provide the guaranteed bit rate, packet delay budget, and packet error rate required for the tele-operation service. The vehicle acts as a User Equipment (UE) with enhanced capabilities for sensor data collection, compression, and uplink transmission. The remote control center, hosting the Tele-Operation Center (TOC), connects to the mobile network as an Application Function (AF) or via a data network. The 5G core network, using the Policy Control Function (PCF) and Session Management Function (SMF), establishes and manages the PDU sessions with the appropriate QoS characteristics for the bidirectional data streams.

Key technical components include precise positioning, which is critical for the remote operator's situational awareness. Specifications like 3GPP TS 38.845 define performance requirements and evaluation for positioning in TOD scenarios. Furthermore, the service relies on end-to-end synchronization and timing, often supported by network-based timing protocols. The data streams are typically handled as separate QoS flows—one for high-bandwidth, low-latency uplink sensor data and another for low-bandwidth, ultra-reliable downlink control commands. Testing and assurance specifications (e.g., TS 37.571) define conformance testing for UE supporting TOD features, ensuring they meet the rigorous performance benchmarks necessary for safe operation.

Purpose & Motivation

Tele-Operated Driving was introduced to address the limitations of fully autonomous vehicles, particularly in complex, unpredictable, or legally constrained driving scenarios. While autonomous driving AI has advanced, there are edge cases—such as construction zones, accident scenes, or system failures—where human judgment and intervention are superior or legally required. TOD provides a fallback solution, allowing a remote operator to safely guide or directly control the vehicle. This enables the deployment of autonomous vehicle services, like robotaxis or autonomous trucks, with a safety net, thereby increasing public trust and regulatory acceptance.

The creation of TOD standards within 3GPP was motivated by the need for a standardized, reliable, and globally interoperable communication framework. Previous ad-hoc solutions using proprietary radio links lacked the scalability, security, and guaranteed performance of a cellular system. 3GPP's work, beginning in later 4G releases and maturing in 5G, aimed to define the precise network requirements—latency under 10-50ms, reliability up to 99.999%, and precise positioning—that mobile operators must meet to support such safety-critical services. It solves the problem of how to extend human driving capability over long distances with near-instantaneous feedback, which is impossible with pre-5G cellular technologies due to their higher latency and less deterministic performance.

Key Features

• Requires Ultra-Reliable Low-Latency Communication (URLLC) with stringent QoS (e.g., <50ms latency, >99.999% reliability)
• Supports bidirectional data flows: uplink for sensor/video data and downlink for control commands
• Often implemented using dedicated 5G network slices to guarantee performance isolation
• Relies on high-accuracy positioning services (e.g., A-GNSS, RTK) for vehicle and context awareness
• Defines specific performance requirements and evaluation methodologies in 3GPP test specs
• Enables remote operation for valet parking, delivery, freight, and passenger vehicle scenarios

Evolution Across Releases

Defining Specifications