Remote train driving from the driver's perspective: insights and design considerations for future control stations

As the rail industry explores remote train operation, understanding the perspectives of train drivers is critical to designing effective remote-control stations. This study investigates how professional train drivers perceive remote train operation compared to traditional in-cab driving and identifies key design requirements for remote operator workstations. Using a mixed-methods approach, 15 licensed train drivers participated in experimental drives of a prototype train under both manual and remote-control conditions. Data was collected through questionnaires and interviews, followed by evaluation of two interface prototypes developed based on participant feedback. Results show that while drivers generally preferred in-cab operation due to better sensory feedback and situational awareness, many recognized the potential of remote operation, particularly if interfaces are ergonomically designed, visually clear, and adaptable to various operational contexts. The findings underscore the importance of human-centered design in developing remote operator workstations that support ergonomic comfort, sensory awareness, and operational efficiency. Simplicity, clarity, and functional relevance emerged as key interface priorities, alongside the need to accommodate differing operational needs between passenger and freight train drivers. These insights provide valuable guidance for the iterative development of safe and effective remote train control systems.

1 Introduction

Rail transport offers several benefits, including reduced traffic congestion, lower pollution levels, safer travel and decreased costs, compared to road transport (Litman, 2008; Wener and Evans, 2011). The growing demand for rail passengers and freight traffic is driving the need for improvements in railway operations. Automation systems, such as Automatic Train Operation (ATO), are seen as a solution. Driverless metros are already being implemented in several cities worldwide. Although research on remote train driving is limited, it could serve as an alternative in some cases or pave the way for fully autonomous trains (Vaidya et al., 2018).

There are several potential benefits to remote train driving, including enabling operators to manage multiple vehicles from a central hub and unlike autonomous driving, which depends on advanced sensors and complex algorithms, remote driving uses simpler systems, such as basic sensors and live video feeds (Zhang, 2020). For trains, this could improve scheduling flexibility, optimize driver efficiency, and reduce the need for drivers at multiple locations. This would increase train frequency and accessibility, promoting rail travel as a sustainable transportation option. Further, remote rail systems are viewed positively by potential users (Cogan et al., 2022), though future implementation plans must address concerns regarding safety, security, and trust (Cogan et al., 2022; Masson et al., 2022; Pacaux-Lemoine et al., 2020).

There are, of course, significant challenges associated with remote train operation. Key obstacles include ensuring robust and uninterrupted connectivity, avoiding latency, addressing cybersecurity threats, and managing the transition of the existing workforce. The importance of collaborative strategies engaging with employees and employee representatives to address concerns about workforce displacement, stakeholder involvement, and the incremental implementation of automation technologies has been emphasized as critical to mitigating risks and maximizing their effectiveness (Morin et al., 2025).

A growing body of research highlights the value of integrating teleoperation alongside the development of automation systems (Ahvenjärvi, 2016; Porathe, 2014). These studies underscore the need for sensors as substitutes for direct sensory input and stress the importance of defining both the type and format of information communicated to remote operators. Moreover, several investigations have pointed to the specific cognitive and operational challenges, such as situational awareness, that teleoperators face in such settings (Kari and Steinert, 2021; Tener and Lanir, 2022). According to Goerke et al. (2024), future train drivers and operators need to be dependable, like traditional train drivers. Moreover, operational monitoring and vigilance become even more important. Essentially, future job profiles for remote and high-speed train drivers have the same requirements including sensory, cognitive, interactive and psychomotor abilities.

The fundamental differences in requirements between freight and passenger train driving include variations in driving speed, number of stops, and the nature of responsibility, whether for passengers or freight. Moreover, freight train drivers often face irregular work schedules, experience greater boredom during operations (Sussman and Coplen, 2000) and encounter a higher incidence of issues related to night operations (Jackson, 2005).

Train driving is primarily a visual task, requiring operators to interpret and integrate a range of visual cues from both internal systems and the external environment (Luke et al., 2006). In remote operation, the lack or distortion of sensory feedback, such as sound or vibration, leads to an overreliance on visual inputs compared to on-board driving (Rybarczyk et al., 2004). To mitigate this, the design of remote operator interfaces must prioritize multimodal information presentation. Wickens' model of human information processing (2008) supports the use of multimodal cues to enhance task performance. Supporting evidence also points to increased operator satisfaction (Triantafyllidis et al., 2020) and improved performance outcomes (Burke et al., 2006) when multimodal feedback is incorporated. Indeed, the loss or alteration of sensory information due to the driver's physical separation from the train remains a central issue in teleoperation research (Kallioniemi et al., 2021).

According to Urassa et al. (2024), performance indicators are crucial for assessing the operational effectiveness of remote railway operations. These include latency, data transfer rate, cybersecurity measures, video quality and camera stability, perception, system integration, permanent connection checks, driver vitality monitoring, and organizational aspects. Because many of these indicators directly involve or affect the driver, it is essential to include drivers in the research and development process.

Consequently, human factors must be a focal point in the development of remote train driving systems. For instance, latency has been shown to increase mental workload and impair performance, though predictive systems, like a display indicating the position as if there was no delay, may help offset these effects (Dybvik et al., 2021). Additionally, understanding the behavioral and experiential differences between remote and on-board train operation remains a critical research gap. Our research group has shown that drivers tend to operate more cautiously in remote conditions and experience higher mental demand, greater effort, and increased frustration compared with driving from the cab (Rosberg et al., 2025). These experiences and the drivers' resulting requirements are what this study aims to address in order to inform the design of remote-operation workstations.

2 Aim and research questions

The aim of this study was to explore train drivers' initial experiences of remote train operation and to gather both their perspectives and suggestions to propose and evaluate two designs of a remote operator workstation. The research questions were:

RQ1: How do train drivers perceive and experience remote train operation compared to traditional in-cab driving?

RQ2: What design elements do train drivers consider most important for a safe, effective, and user-friendly remote operator workstation?

3 Method

This study employs a mixed-methods approach, incorporating two experimental drives, one using manual operation and the other remote train operation, and combines questionnaires and interviews conducted both online and at the Training and Education Centre in Ängelholm. The methods and procedures have been approved by the Swedish Ethical Review Authority (Dnr 2025-01715-01). For RQ1, 15 expert train drivers got to experience two experimental drives of a model train: one with manual and one with remote train operation. This was followed by a questionnaire with open-ended questions as well as interviews after both driving conditions were used. For RQ2, the same approach was used, plus a second round of a questionnaire to evaluate design suggestions presented to them (see Figure 1).

Figure 1

Figure 1. Flowchart of study and analysis procedure.

3.1 Participants

Participants were recruited via social media (Facebook) and invitations from the Train Driver Education School in Ängelholm to former students. The inclusion criterion was possession of a valid train driver's license; participants included both passenger and freight train drivers. A total of 15 participants took part in the study, comprising 1 female and 14 males, aged between 25 and 61 years. Participants received oral and written information about the study, were informed of their right to withdraw before signing consent, and were assigned a unique ID. In the second round of the questionnaire, aimed at evaluating the design suggestions, the same 15 participants were invited, but only 8 volunteered to take part.

3.2 Equipment

The battery-powered train was developed by engineering students at Linköping University, with a maximum speed of 35 km/h. However the drivers were driving on-sight, meaning that they were relying only on the sight distance and not protected by any train protection system. Under these preconditions the maximum speed was 25 km/h (Figure 2). The Swedish National Road and transport Research Institute (VTI) train simulator (Rosberg and Thorslund, 2025) serves as the remote-control station. The setup was developed with a focus on video quality and control mechanisms. A full HD main camera provided a clear forward view for the remote driver. A full HD main camera provided a clear forward view for the remote driver. In both driving conditions, remote and in-cab, the train was controlled by one analog driving stick, enabling traction and brake. In the DMI speed was displayed and a horn function provided. Voysys video software was used to generate and decode the video stream, which was transmitted via a 5G modem to the control room. Key objectives included low latency, where measurements showed < 150 ms, and minimal blurriness. In the control room setup, the internet was connected with a fixed cable, via a firewall. The train control channel was secured by a VPN tunnel.

Figure 2

Figure 2. The remote-driving set-up with vehicle and control station.

3.3 Procedure

Each participant completed two runs on a 700-meter track, one under remote driving and one under traditional operation, with counterbalanced order. The track included two road crossings and one pedestrian crossing. After each run, participants answered four open-ended questions on driver experience, preferences and design suggestions: (A) How would you describe the difference between driving the train from the driver's cab vs. the remote-control station? (B) What do you think is most important to consider when developing remotely controlled trains? (C) What is important to consider when designing the remote-control station? (D) Do you prefer driving from the train or from the remote-control station? In addition, after the second run, an interview with the participants was conducted with a focus on requirements for usability, safety, ergonomics, and efficiency when designing a remote operator workstation. The eight interview questions are presented in Appendix 1.

Based on input from the questionnaire and interviews, two initial interface prototypes were developed by cognitive science students at Linköping University. The designs were created using Figma and shared online with the participants for evaluation. To streamline the evaluation process and minimize the time required from participants, the assessment was conducted using structured questions with predefined answer options (see Appendix 3). The procedure is displayed in Figure 1.

3.4 Thematic analysis

The interviews were transcribed using Microsoft Word and subsequently translated into English using ChatGPT, as were the responses to the open-ended questions. Following Braun and Clarke (2006) six-step process for inductive analysis, qualitative data from both the open-ended questions and participant interviews was used to identify, analyze, and report patterns (themes). This approach was chosen to systematically capture the key meanings and insights from the two drives, focusing on operators' perceptions, experiences, preferences, and requirements regarding remote train control workstations. After familiarization with the data, initial codes were generated, which were then grouped into broader themes, see subsequent sections for questionnaire and interviews. These themes were then refined and validated by checking them against the dataset and connecting participants'responses to each theme. Minor adjustments of the names were made resulting in those themes presented in the results section.

3.5 Questionnaire

The initial codes were generated by a single coder (first author) based on the questionnaire responses, with iterative refinement and detailed documentation used to reduce potential coder bias. In addition, the co-authors reviewed and challenged emerging analytic decisions to enhance the credibility and trustworthiness of the analysis. The following initial codes were generated: “Better visibility”, “Loss of control”, “Video quality”, “Interface”, “Feedback”, “Visual input”, “Attitude toward remote operation” and “Safety concerns”. These were grouped into 6 themes and core meanings presented in Table 1.

Table 1

Table 1. Main themes, core meanings and participant expressing topics within this theme.

3.6 Interviews

The initial codes generated by a single coder based on the interviews were reviewed and grouped into 8 broader thematic categories reflecting shared concepts and patterns across the dataset, such as “workstation comfort,” “sensory feedback,” and “emergency handling”, see Appendix 2. The themes were then refined and defined to capture the core meanings and relationships in the data, which informed the final analysis and interpretation. These broader themes and relations are presented in Table 2.

Table 2

Table 2. Themes, core meanings, related questions, and participant expressing topics within this theme.

4 Results

Results are presented separately for the questionnaire, interviews and the evaluation of prototypes.

4.1 Thematic analysis questionnaire

The thematic analysis of open-ended questions identified six major themes related to the perception of driving remotely and the design of operator workstations for remote train control.

4.1.1 Theme 1: feeling of control and situational awareness

Many participants emphasized a greater feeling of control and better situational awareness when operating the train from the physical cab. Participants described difficulties in estimating speed, braking distances, and track conditions when driving remotely. The physical presence in the cab contributed to their perception of safety and precision.

“It was harder to judge speed from the control station. I felt more in control from the cab.” (P2)

“Braking feel was significantly better when physically present in the cab.” (P10)

4.1.2 Theme 2: sensory feedback and physical experience

Participants noted that the lack of sensory cues, such as vibrations, motion, and engine noise, during remote operation diminished their driving experience and connection with the train. These physical sensations were considered essential for maintaining awareness and reacting to the train's behavior in real time.

“The biggest difference is the absence of external stimuli like sounds, vibrations...” (P1)

“Depth perception and side vision were missing, which made me feel unsure.” (P14)

4.1.3 Theme 3: safety concerns

Safety emerged as a prominent theme, with several participants expressing concerns about operating trains remotely. The inability to react intuitively to on-site conditions, respond to emergencies, or maintain awareness of technical malfunctions were highlighted as potential risks.

“Remote control requires more effort and reduces safety.” (P3)

“What happens if the camera feed fails? Who takes responsibility for passengers?” (P15)

4.1.4 Theme 4: interface and environment design

Participants stressed the importance of designing the remote-control station to resemble a real cab as closely as possible, especially in terms of control layout and visual orientation. Additionally, ergonomic seating, minimal distractions, and spaciousness were emphasized as critical to operator focus and comfort.

“Familiarity helps—make it look like the real cab.” (P2)

“It should be a calm, spacious environment with a comfortable seat.” (P13)

4.1.5 Theme 5: visual input and camera technology

High-quality, low-latency video feeds and peripheral vision capabilities were seen as essential for safe and confident remote operation. Participants frequently mentioned that the current video setup lacked sufficient depth and breadth of view.

“Important to have stable and high-quality video. I couldn't see well through the current camera.” (P4)

“I want to be able to see in all directions, not just straight ahead.” (P14)

4.1.6 Theme 6: openness to remote operation

While most participants preferred driving from the cab, several expressed openness to remote operation under certain conditions. For example, remote control could be acceptable for secured routes, considered safe, predictable, and controlled, with minimal risk factors, or if the technology significantly improves.

“From the train, definitely. But I'm not against remote control if it's safe and well developed.” (P4)

“I could imagine using remote for secured movements.” (P15)

4.1.7 Summary

In summary, participants favored driving from the cab due to enhanced control, sensory feedback, and safety. However, with improvements in technology, particularly in visual feedback and interface design, some were open to the future potential of remote operation. These findings underscore the importance of human-centered design in remote train control systems and the critical role of sensory and situational feedback in train operator performance.

4.2 Thematic analysis interviews

The thematic analysis of interviews identified four major themes related to the design and operation of operator workstations for remote train control.

4.2.1 Theme 1: ergonomic and sensory environment

Participants emphasized that the workstation environment should support operator comfort and replicate the sensory feedback of in-cab driving as closely as possible. Ergonomic seating, noise reduction, and climate control were frequently mentioned as crucial factors to maintain operator focus and reduce fatigue during long shifts. One participant stated:

“It is important to have a calm, quiet environment with good seating to stay alert. “ (P8).

4.2.2 Theme 2: challenges in the transition to remote control

A significant concern among participants was the challenge of adapting from direct in-cab control to remote operation, particularly regarding safety and situational awareness. The loss of tactile and auditory cues was seen as a potential risk factor, complicating tasks such as speed regulation and response to unexpected events. Participants highlighted the importance of workstation design in mitigating these challenges by providing clear, intuitive feedback and tools that help operators anticipate and react promptly. As one participant explained:

“Without the vibrations and engine sounds, it's harder to judge speed and when you need to slow down. The workstation needs to give clear signals that replace what you usually feel.” (P9)

The absence of physical sensations like vibrations and sounds was perceived as a major limitation that could increase mental strain. One participant stated:

“The mental effort increases when you're remote because you don't get the usual physical cues, like the feeling of the train moving. “ (P8)

Several participants also noted the need for sensory feedback mechanisms (e.g., haptic feedback or sounds) to compensate for the lack of direct physical interaction with the train controls.

4.2.3 Theme 3: essential workstation features and tools

Participants identified several features they consider indispensable for effective remote operation. High-quality, multi-angle camera views with zoom functionality were seen as essential for monitoring track conditions and surroundings. Integrated digital maps and real-time updates on train status, including speed, signals, and obstacles, were also highlighted as critical. The interface should provide clear, intuitive feedback that allows operators to maintain situational awareness without being overloaded with information. Participants suggested customizable displays and alerts that can be tailored to individual preferences. One participant commented:

“Having multiple camera angles and the ability to zoom in on areas is a must. Plus, an integrated map that shows signals and upcoming obstacles helps keep everything in view.” (P12)

4.2.4 Theme 4: handling emergency situations

All participants agreed that the workstation must include reliable and immediate options for managing emergencies. This includes easy access to emergency stop functions, clear signaling tools to alert other personnel, and protocols for requesting assistance. Some participants suggested that emergency functions should be physically distinct and fail-safe to avoid accidental activation or delay in critical situations. For example:

“In an emergency, you need to be able to stop the train instantly and signal for help without confusion. The controls for this must be obvious and easy to reach.” (P5)

4.2.5 Summary

Overall, the participants' insights underscore the importance of designing remote operator workstations that combine ergonomic comfort with advanced sensory and informational feedback to support safety, efficiency, and operator well-being during remote train control.

4.3 Evaluation of prototypes

Building on the insights from both the questionnaire and interview analyses, several key design implications emerged for the development of remote train operator workstations. The findings consistently emphasized the need for enhanced sensory feedback, ergonomic comfort, intuitive control layouts, and improved visual coverage to support situational awareness and safety during remote operation. These insights directly informed the design and evaluation of prototype workstation concepts. Given that the subsequent evaluation was conducted through an online survey, the design suggestions were limited to features that could be effectively represented and assessed visually. Consequently, two design proposals were developed, each integrating different approaches to visual display, control placement, and information accessibility.

See Figures 3, 4 for the two initial prototypes designed based on the results from questionnaires and interviews. In suggestion 1 rear-view mirrors, speedometer, and GPS have been integrated on the main screen and the iPad control panel is in original below the screen, except from the speedometer which has been moved to the main screen. I suggestion 2, two extra screens are added next to the main screen for a wider view, rear-view mirrors are added next to the control panel, while a speedometer and GPS are integrated on the control panel. Of the 15 participants involved in the test-drive and interviews, 8 participated in the evaluation of the prototypes. Five of the participants preferred design suggestion 2 prior to alternative 1. The most appreciated features of design 1 were camera angles (n = 3), GPS-positioning (n = 4), rear view mirror positioning (n = 3), speedometer positioning (n = 4), and design 2 were the amount of camera views (n = 4), camera angles (n = 3), speedometer positioning (n = 3). The open-ended questions were answered by 7 participants and the responses revealed several recurring themes:

Figure 3

Figure 3. Design suggestion 1 with rear-view mirrors, speedometer, and GPS integrated on the main screen (above) and an iPad control panel (below).

Figure 4

Figure 4. Design suggestion 2 with extra screens for a wider view (above) and rear-view mirrors next to the control panel (below). Speedometer and GPS are integrated on the control panel.

4.3.1 Theme 1: perceived clutter and information overload

Several participants expressed that Design 2 appeared too cluttered, citing an overabundance of screens and a control panel that was “too rich in content.” The visual complexity was seen as a potential distraction during operation (P1, P3, P5, P8).

4.3.2 Theme 2: mixed views on GPS utility

Some participants questioned the need for a GPS display, arguing that experienced drivers possess adequate line knowledge and access to line manuals, and that the GPS could become a distraction (P2, P6). Others, particularly those with freight train experience, emphasized the critical importance of knowing one's exact position to adjust traction and avoid getting stuck on slopes (P7). One participant preferred a hybrid solution, incorporating GPS and speed in a central view (P3).

4.3.3 Theme 3: camera views and visual field

Several respondents appreciated wide or angled views, which they considered helpful for track visibility and awareness of other trains. However, some questioned the necessity of additional side views (P2, P5, P6, P8).

4.3.4 Theme 4: analog vs. digital controls

One participant strongly advocated for the use of analog buttons instead of digital functions, arguing that analog components are more robust, less prone to failure in extreme temperatures, and easier to manage under pressure (P6).

4.3.5 Theme 5: need for critical operational information

The inclusion of certain key operational indicators, such as real-time information about overhead line voltage, was highlighted as essential but lacking in both prototypes (P6).

4.3.6 Theme 6: design preferences

A few participants favored a combination of elements from both prototypes. One explicitly suggested merging the broader layout of prototype 2 with the central GPS and speed display features in prototype 1 (P3). Figure 5 illustrates this proposed combination.

Figure 5

Figure 5. A combination of the two prototypes was suggested, merging the broader layout of prototype 2 with the central GPS and speed display features in prototype 1.

Overall, the evaluation responses underscored the importance of simplicity, clarity, and operational relevance in the interface design. The findings also illustrate the diverse preferences and operational needs between different types of train operators, particularly between passenger and freight services.

5 Discussion

The study explored train drivers' perceptions of remote train operation and their preferences for remote operator workstation design. Drawing from test-driving sessions, questionnaires, interviews, and prototype evaluations, several key themes emerged related to operator experience, safety, interface design, and situational awareness.

5.1 Perception of remote train operation

Consistent with our prior findings that drivers operate more cautiously and experience higher mental workload and frustration in remote settings (Rosberg et al., 2025), participants in this study expressed similar concerns regarding the sensory and cognitive challenges of remote train operation. The findings revealed that while participants generally favored traditional in-cab driving due to its direct control, sensory input, and situational awareness, some expressed openness to the idea of remote train operation, provided certain technological and interface limitations were addressed. These mixed perspectives align with previous research indicating that human factors such as sensory feedback and situational awareness are critical to train operation performance (Luke et al., 2006; Rybarczyk et al., 2004).

Participants highlighted concerns about the lack of sensory immersion and physical presence when driving remotely, which supports earlier studies emphasizing the over-reliance on visual modalities in remote settings and the need for multimodal feedback (Kallioniemi et al., 2021; Wickens, 2008). In particular, latency and the absence of tactile and auditory cues were perceived as potential risks, echoing the concerns raised by Dybvik et al. (2021) regarding increased cognitive load and performance decline under such conditions. These findings further support the performance indicators identified by Urassa et al. (2024), who emphasize latency, perception accuracy, video quality, and system integration as crucial factors for maintaining operational effectiveness in remote railway contexts. The concerns expressed by drivers in this study demonstrate how closely these technical factors are intertwined with human performance and safety.

However, there were also clear indications of the perceived potential of remote train driving, particularly in terms of flexibility, operational efficiency, and future integration with automation. This supports arguments by Zhang (2020) and Cogan et al. (2022) about the feasibility and public acceptance of remote systems. Several participants envisioned remote operation as a realistic future alternative, especially if designed in a way that replicates key elements of the on-board experience and ensures safety and reliability.

5.2 Important design considerations

The insights from both the interviews and prototype evaluations highlighted several critical design priorities for remote train operator workstations. Across participants, there was a strong and consistent preference for clean, intuitive interfaces that present essential operational data, such as speed, braking status, signals, and overhead line voltage, without causing visual overload. Unnecessary elements, such as cluttered maps or overly detailed control panels, were seen as distracting and counterproductive. This aligns with research advocating for ergonomic, user-centered design that reduces cognitive load and supports efficient task execution (Wickens, 2008; Triantafyllidis et al., 2020).

To compensate for the lack of physical presence inherent in remote operation, participants emphasized the need for high-quality, wide-angle video feeds, ideally with options for zooming or toggling between multiple views. This aligns with Urassa et al. (2024), who identify video quality, camera stability, and reliable system integration as key indicators of successful remote operation performance. The design preferences expressed by operators reflect the operational importance of these factors and highlight the risks introduced when they are insufficiently addressed. These features were considered especially important when approaching platforms, detecting obstacles, or monitoring the rear of the train. Additionally, several participants mentioned the potential value of auditory or haptic feedback to enhance situational awareness and immersion.

A notable finding was the divergence in design preferences between freight and passenger train operators. Freight drivers strongly emphasized the importance of positional awareness tools like GPS, which they rely on for traction optimization and stall prevention, particularly on hilly or rural lines. In contrast, some passenger train operators considered GPS redundant, relying more on their line knowledge and printed manuals. This disparity highlights the need for customizable or adaptable interfaces that can accommodate different operational demands and use contexts.

Concerns about system reliability in extreme environments were also prominent. Some participants expressed a preference for physical buttons and analog controls over digital touchscreens, which were viewed as more vulnerable to failure under heat, cold, or other challenging conditions, such as direct sun glare on digital screens. This suggests that hybrid interfaces, combining digital displays with tactile elements, can enhance both usability and system redundancy. Such approaches have long proven effective in automotive design (Murali et al., 2022).

Finally, participants stressed the importance of ergonomic workstation design, including screen placement, adjustable seating, posture support, and features that reduce fatigue over extended sessions. These considerations are crucial for maintaining operator vigilance, safety, and well-being, particularly in remote contexts where monotony and mental workload can fluctuate. Such concerns align with Goerke et al. (2024, 2025) findings on inclusive and accessible design for future train operator roles.

5.3 Methods discussion and future work

A major strength of this study is its use of multiple methods to gather and triangulate data, enhancing the credibility and depth of the findings. The integration of interview insights, real-time interaction with a remote driving setup, and interface evaluation enabled a comprehensive understanding of operator perspectives. The use of thematic analysis further supported systematic and transparent interpretation of qualitative data, making it possible to identify patterns in user needs and concerns.

However, a key limitation is the relatively small number of participants, particularly in the prototype evaluation phase. While the responses were rich and informative, a larger sample could have increased the generalizability of the findings. Additionally, as participation in the study was voluntary, there may have been a selection bias favoring individuals with a stronger interest in technology or innovation. Future train drivers, often part of the so-called “gaming generation,” are likely to be even more technologically inclined and may therefore require different interface designs. Furthermore, the current participants were accustomed to manual setups, which may have limited their ability to envision alternative configurations more suitable for remote operation. Despite these limitations, the findings offer valuable insights into the design of human-centered remote train operation systems and highlight critical user requirements that could inform future research and system development.

Given the exploratory nature and the limited number of participants in this study, future research should aim to expand the sample size and include a broader demographic of train operators, such as those from different countries, operating conditions, and levels of experience with digital tools. This would improve the generalizability of findings and capture additional variations in operator needs and expectations. Also, in this study the prototype evaluation was made descriptively, and a future study could include a more systematic evaluation with using usability scales to strengthen the claims. Moreover, future studies should involve longitudinal testing of remote driving interfaces under more realistic operational conditions. While this study included prototype evaluations and simulated driving, extended use over time would offer deeper insight into factors like fatigue, situational awareness, and interface learning curves.

Further development and testing of multimodal feedback systems (e.g., haptic, audio, or environmental cues) could also be explored. These features were frequently mentioned by participants as important in compensating for the lack of physical presence in the cab, especially for enhancing situational awareness and response to critical events. It would also be beneficial to involve ergonomics specialists and UX designers in co-creation sessions with train drivers to refine interface design iteratively. A participatory design approach could ensure that future workstations are truly adapted to the practical and cognitive demands of remote train operation. Finally, as the transition to remote operation also raises new demands on organizational routines, roles, and training, socio-technical studies could complement interface development by exploring how remote operation integrates with broader railway systems and human workflows. Physical abilities are not important for these job profiles and practically, that underlines that the job of future train operators could be designed as a completely accessible workspace (Goerke et al., 2024, 2025). In line with this, future studies can adopt an even broader and inclusive approach to gather diverse experiences and ideas from train drivers.

6 Conclusion

The results from this mixed methods study of user experience and design evaluation revealed that while participants generally favored in-cab driving for its superior sensory control and situational awareness, many acknowledged the future potential of remote operation, provided that technical systems and human-centered design are meaningfully improved. Across questionnaire, interviews and prototype evaluations, participants consistently emphasized that remote operator workstations must be simple, functional, and adaptable to different operational roles. Key design principles included minimizing visual clutter, providing reliable sensory feedback, and ensuring ergonomic support to sustain vigilance and comfort. Although the prototype evaluation was conducted over a short 700-meter course at low operating speeds, the insights gathered offer an important indication of how technical and interface factors shape driver acceptance and perceived feasibility of remote train operation. Future work in more varied operational conditions will further validate and extend these findings. Overall, the study contributes actionable guidance for the iterative development of safe, efficient, and user-responsive remote train control systems and underscores the value of involving end-users throughout the design process.

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics statement

The studies involving humans were approved by Etikprövingsmyndigheten Sverige. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

Author contributions

BT: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Visualization, Writing – original draft, Writing – review & editing. FB: Methodology, Writing – review & editing. TR: Conceptualization, Methodology, Software, Writing – review & editing.

Funding

The author(s) declare that financial support was received for the research and/or publication of this article. This project was funded by a grant from Linköping University under the “Verification for Utilization” program within the Transport and Mobility profile. This work was possible thanks to a synergy with the ongoing activities of EURail FP6 Future Project.

Acknowledgments

The authors would like to express their gratitude to the Training and Education Centre in Ängelholm for providing access to their facilities and assisting with participant recruitment, to the participants for their valuable contributions, and to the students from Linköping University for their work in developing the design prototypes.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declare that Gen AI was used in the creation of this manuscript. Chat GPT was used for translation of the transcribed interviews.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Ahvenjärvi, S. (2016). The human element and autonomous ships. TransNav Int. J. Mar. Navig. Saf. Sea Transp. 10, 517–521. doi: 10.12716/1001.10.03.18

Crossref Full Text | Google Scholar

Braun, V., and Clarke, V. (2006). Using thematic analysis in psychology. Qual. Res. Psychol. 3, 77–101. doi: 10.1191/1478088706qp063oa

Crossref Full Text | Google Scholar

Burke, J. L., Prewett, M. S., Gray, A. A., Yang, L., Stilson, F. R., Coovert, M. D., et al. (2006). “Comparing the effects of visual-auditory and visual-tactile feedback on user performance: a meta-analysis,” in Proceedings of the 8th International Conference on Multimodal Interfaces (Banff, AB), 108–117.

Google Scholar

Cogan, B., Tandetzki, J., and Milius, B. (2022). Passenger acceptability of teleoperation in railways. Future Transp. 2, 956–969. doi: 10.3390/futuretransp2040053

Crossref Full Text | Google Scholar

Dybvik, H., Løland, M., Gerstenberg, A., Bjørnerud Slåttsveen, K., and Steinert, M. (2021). A low-cost predictive display for teleoperation: investigating effects on human performance and workload. Int. J. Hum.-Comput. Stud. 145:102536. doi: 10.1016/j.ijhcs.2020.102536

Crossref Full Text | Google Scholar

Goerke, P., Maier, J., Zierke, O., Marggraf-Micheel, C., Brandenburger, N., and Naumann, A. (2024). “Job requirements for future train drivers: a job analysis,” in Human Factors and Ergonomics Society Europe Chapter 2024 Annual Conference (Lübeck).

Google Scholar

Goerke, P., Zierke, O., Maier, J., Marggraf-Micheel, C., Naumann, A., and Brandenburger, N. (2025). “What requirements do future train drivers need to fulfill? A comparison of job profiles,” in 6th German Conference on Rail Human Factors (Berlin).

Google Scholar

Jackson, C. (2005). Effective and speedy WIN the traffic. Railw. Gaz. Int. 161, 267.

Google Scholar

Kallioniemi, P., Burova, A., Mäkelä, J., Keskinen, T., Ronkainen, K., Mäkelä, V., et al. (2021). Multimodal warnings in remote operation: the case study on remote driving. Multimodal Technol. Interact. 5:44. doi: 10.3390/mti5080044

Crossref Full Text | Google Scholar

Kari, R., and Steinert, M. (2021). Human factor issues in remote ship operations: lesson learned by studying different domains. J. Mar. Sci. Eng. 9:385. doi: 10.3390/jmse9040385

Crossref Full Text | Google Scholar

Litman, T. (2008). Rail transit in America: a comprehensive evaluation of benefits. J. Public Transp. 11:2.

Google Scholar

Luke, T., Brook-Carter, N., Parkes, A. M., Grimes, E., and Mills, A. (2006). An investigation of train driver visual strategies. Cogn. Technol. Work 8, 15–29. doi: 10.1007/s10111-005-0015-7

Crossref Full Text | Google Scholar

Masson, É., Richard, P., Boussif, A., Gadmer, Q., David, P., Robert, E., et al. (2022). “TC-rail: railways remote driving further experiments,” in 13th World Congress on Railway Research (WCRR), 6–10 June 2022 (Birmingham).

Google Scholar

Morin, X., Olsson, N. O. E., and Lau, A. (2025). Expected challenges and anticipated benefits of implementing remote train control and automatic train operation: a tramway case study. Future Transp. 5:73. doi: 10.3390/futuretransp5020073

Crossref Full Text | Google Scholar

Murali, P. K., Kaboli, M., and Dahiya, R. (2022). Intelligent in-vehicle interaction technologies. Adv. Intell. Syst. 4:2100122. doi: 10.1002/aisy.202100122

Crossref Full Text | Google Scholar

Pacaux-Lemoine, M.-P., Gadmer, Q., and Richard, P. (2020). “Train remote driving: a human-machine cooperation point of view,” in Proceedings of the 2020 IEEE International Conference on Human-Machine Systems (ICHMS) (Rome), 1–4.

Google Scholar

Porathe, T. (2014). “Remote monitoring and control of unmanned vessels: the MUNIN shore control centre,” in Proceedings of the 13th International Conference on Computer Applications and Information Technology in the Maritime Industrie (COMPIT 2014) (Redworth), 460–467.

Google Scholar

Rosberg, T., and Thorslund, B. (2025). Effects of European rail traffic management system onboard version on running time, driver target braking, and driver workload. Transp. Res. Rec. 2679, 414–425. doi: 10.1177/03611981241311558

Crossref Full Text | Google Scholar

Rosberg, T., Turén, F., Urassa, P., and Thorslund, B. (2025). “Comparing remote and traditional train driving: effects on behavior, perspective, and workload,” in Proceedings of SweCog 2025 (Lund).

Google Scholar

Rybarczyk, Y., Mestre, D., Hoppenot, P., and Colle, E. (2004). A biological model for the evaluation of human-machine adaptation. AMSE Model. Meas. Control 65, 23–33.

Google Scholar

Sussman, D., and Coplen, M. (2000). Fatigue and alertness in the United States railroad industry part 1: the nature of the problem. Transp. Res. Part F Traffic Psychol. Behav. 3, 211–220. doi: 10.1016/S1369-8478(01)00005-5

Crossref Full Text | Google Scholar

Tener, F., and Lanir, J. (2022). “Driving from a distance: challenges and guidelines for autonomous vehicle teleoperation interfaces,” in Proceedings of the 2022 CHI Conference on Human Factors in Computing Systems (New Orleans, LA), 1–13.

Google Scholar

Triantafyllidis, E., Mcgreavy, C., Gu, J., and Li, Z. (2020). Study of multimodal interfaces and the improvements on teleoperation. IEEE Access 8, 78213–78227. doi: 10.1109/ACCESS.2020.2990080

Crossref Full Text | Google Scholar

Urassa, P., Olsson, N. O. E., Lau, A., Mandhaniya, P., and Andersen, B. (2024). Performance measurement in railway remote driving implementations. IET Intell. Transp. Syst. 18, 2759–2774. doi: 10.1049/itr2.12580

Crossref Full Text | Google Scholar

Vaidya, S., Ambad, P., and Bhosle, S. (2018). Industry 4.0–a glimpse. Procedia Manuf. 20, 233–238. doi: 10.1016/j.promfg.2018.02.034

Crossref Full Text | Google Scholar

Wener, R. E., and Evans, G. W. (2011). Comparing stress of car and train commuters. Transp. Res. Part F Traffic Psychol. Behav. 14, 111–116. doi: 10.1016/j.trf.2010.11.008

Crossref Full Text | Google Scholar

Wickens, C. D. (2008). Multiple resources and mental workload. Hum. Factors 50, 449–455. doi: 10.1518/001872008X288394

PubMed Abstract | Crossref Full Text | Google Scholar

Zhang, T. (2020). Toward automated vehicle teleoperation: vision, opportunities, and challenges. IEEE Internet Things J. 7, 11099–11107. doi: 10.1109/JIOT.2020.3028766

Crossref Full Text | Google Scholar

Appendices

Appendix 1. Interview questions

1. What do you consider most important when designing an operator workstation for remote train control?

2. What challenges do you anticipate in the transition from in-cab driving to remote control, and how can the workstation's design help address these challenges?

3. How should the operator workstation be designed to ensure that the operator maintains focus and avoids fatigue during long periods of remote operation?

4. Which specific tools or features do you think would be most useful at the operator workstation, such as camera views, zoom functions, or integrated maps?

5. What are your thoughts on the interface regarding clear and intuitive feedback on train status, for example speed, signal changes, and upcoming obstacles?

6. How should emergency situations be handled, such as signaling for assistance, stopping the train, or warning other personnel?

7. How do you think the same level of control as in-cab operation can be achieved, especially when handling unexpected events?

8. What types of feedback (e.g., vibrations, sounds, or haptic feedback) do you think would be helpful during remote operation?

Appendix 2. Initial codes from interviews

Table A1

Table A1. Theme areas and initial codes.

Appendix 3. On-line questionnaire

Table A2

Table A2. Questions and predefined answer options for the evaluation.