Autonomous Control in Military Logistics Vehicles: Trust and Safety Analysis

Abstract

Ground vehicles are increasingly designed to incorporate autonomous control for better performance, control and efficiency. Such control is particularly critical for military logistics vehicles where drivers are carrying sensitive loads through potentially threatening routes. It is imperative therefore to evaluate what role does autonomy play to help safety, and whether drivers trust autonomous control. In this paper we investigate the use of semiautonomous vehicles used for military logistics and carry out human factors analysis to reflect on trust and safety issues that emerge from the driving of such vehicles.

Impact of Automation on Drivers’ Performance in Agricultural Semi-Autonomous Vehicles

“…The VSTA effect was not significant, but in the case of ICMTA, increasing automation level was associated with lower reaction time and number of errors. This result confirms the findings of previous studies (Gempton et al, 2013;Sethumadhavan, 2009) by presenting automation benefits on drivers' performance.…”

“…Degree of automation, task difficulty, and task type can directly affect these components. Improvement in reaction time was reported when drivers were driving with an autonomous control mode in military semi-autonomous vehicles (Gempton et al, 2013). Another experiment by Sethumadhavan (2009) demonstrated the benefits of high levels of automation in multi-task environments where operators had to perform multiple tasks concurrently.…”

Handing over the Keys: A Qualitative Study of the Experience of Automation in Driving

“…Firstly, problems were identified in the process of handing control from the machine back to the driver. These results reflect the findings of Gempton et al (2013) work with automated military logistic vehicles and also Helldin et al, (2013) on people's ability to recognize problems with automated systems.…”

“…It is notable that the majority of this work has studied Assisted Cruise Control (ACC), as they are currently the most widely commercially available systems. Human factors research is typically founded on cognitive psychology principles and investigates questions surrounding the impact of automated systems on drivers' cognitive load (Brookhuis et al, 2009), attention (Young, Birrell & Stanton, 2011), and safety (Gempton, Skalistis, Furness, Shaikh & Petrovic, 2013), as well as the ability of humans to recognise when there are problems with the technology (Helldin, Falkman, Rivviero & Davidsson, 2013) or the likelihood of the technology to interrupt driving (Stanton & Young, 2005). Findings of this research suggest that ACC can lead to lower max speeds and less breaking of speed limits (Vollrath, Schleicher & Gelau, 2011), less cognitive load on drivers (Stanton & Young, 2005), less stress, and less fatigue (Neubauer, Matthews, Langheim, Saxby, 2012).…”

“…However, studies of automation in cars have also demonstrated undesirable findings. For example, increasing automation does not necessarily reduce human error (Gempton, Skalistis, Furness, Shaikh & Petrovic, 2013). Automation has been found to undermine driver engagement in the driving task (Neubauer, Matthews, Langheim & Saxby, 2012;Stanton & Young, 2005;Jamson, Merat, Carsten & Lai, 2013), to cause slower response times (Vollrath, Schleicher & Gelau, 2011), to increase stress and to induce fatigue (Greenlee, DeLucia & Newton, 2018;Neubauer, Matthews, Langheim & Saxby, 2012).…”