# Safety, Efficiency, and Mental Workload of Predictive Display in Simulated Teledriving

## TL;DR

This study examines if predictive displays help remote drivers maintain safety and efficiency under typical network delays in urban driving simulations.

## Contribution

The study evaluates the effectiveness of predictive displays in teledriving under realistic network delays for the first time.

## Key findings

- Participants tolerated 150 ms delays without significant performance degradation.
- Predictive displays did not improve performance or reduce mental workload.
- Crash and error rates remained low across all delay conditions.

## Abstract

Vehicle remote driving services are increasingly used in urban settings. Yet, vehicle-operator communication time delays may pose a challenge for teleoperators in maintaining safety and efficiency. The purpose of this study was to examine whether Predictive Displays (PDs), which show the vehicle’s predicted real-time position, improve performance, safety, and mental workload under moderate time delays typical of 4G/5G networks. Twenty-nine participants drove a simulated urban route containing pedestrian crossings, overtaking, gap acceptance, and traffic light challenges under three conditions: 50 ms delay (baseline), 150 ms delay without PD, and 150 ms delay with PD. We analyzed the counts of crashes and navigation errors, task completion times, and the probability and intensity of braking and steering events, as well as self-reports of workload and usability. Results indicate that though descriptive trends indicated slightly sharper steering and braking under the 150 ms time delay conditions, the 150 ms time delay did not significantly degrade performance or increase workload compared with the 50 ms baseline. In addition, the PD neither improved performance nor reduced workload. Overall, participants demonstrated tolerance to typical 4G/5G network time delays, leaving little room for improvement rendering the necessitating of PDs.