Partially Automated Research Articles

Who is performing the driving tasks after interventions? Investigating drivers' understanding of mode transition logic in automated vehicles

Mode awareness is important for the safe use of automated vehicles, yet drivers' understanding of mode transitions has not been sufficiently investigated. In this study, we administered an online survey to 838 respondents to examine their understanding of control responsibilities in partial and conditional driving automation with four types of interventions (brake pedal, steering wheel, gas pedal, and take-over request). Results show that most drivers understand that they are responsible for speed and distance control after brake pedal interventions and steering control after steering wheel interventions. However, drivers have mixed responses regarding the responsibility for speed and distance control after steering wheel interventions and the responsibility for steering control after gas pedal interventions. With a higher automation level (conditional driving automation), drivers expect automation to remain responsible more often compared to a lower automation level (partial driving automation). Regarding Hands-on requirements, more than 99% of respondents answered that drivers would keep their hands on the steering wheel after all intervention types in partial automation, while 60–95% would place their hands on the wheel after various intervention types in conditional automation. A misalignment between actual logic and drivers' expectations regarding control responsibilities is observed by comparing survey responses to the mode transition logic of commercial partially automated vehicles. To resolve confusion about control responsibilities and ensure consistent expectations, we propose implementing a consistent mode design and providing enhanced information to drivers.

Combining optimization and dynamic movement primitives for planning energy optimal forestry crane motions

Forestry cranes are an important tool for safe and efficient timber harvesting with forestry machines. However, their complex manual control often led to inefficiencies and excessive energy usage, due to the many joysticks and buttons that must be used in a precise sequence to perform efficient movements. To address this, the industry is increasingly turning to partial automation, making manual control more intuitive for the operator and, consequently, achieving improvements in energy efficiency. This article introduces a novel approach to energy-optimal motion planning that can be used along with a feedback control system to automate crane motions, taking over portions of the operator’s work. Our method combines dynamic movement primitives (DMPs) and an energy-optimization algorithm. DMPs is a machine learning technique for motion planning based on human demonstrations, while the optimization algorithm exploits the crane’s redundancy to find energy-optimal trajectories. Simulation results show that DMPs can replicate human-like controlled motions with a 25% reduction in energy consumption. However, our energy optimization algorithm shows improvements of over 40%, providing substantial energy savings and a promising pathway towards environmentally friendly partially automated machines.

Impact of Partially Automated Driving Functions on Forensic Accident Reconstruction: A Simulator Study on Driver Reaction Behavior in the Event of a Malfunctioning System Behavior.

Partially automated driving functions (SAE Level 2) can control a vehicle's longitudinal and lateral movements. However, taking over the driving task involves automation risks that the driver must manage. In severe accidents, the driver's ability to avoid a collision must be assessed, considering their expected reaction behavior. The primary goal of this study is to generate essential data on driver reaction behavior in case of malfunctions in partially automated driving functions for use in legal affairs. A simulator study with two scenarios involving 32 subjects was conducted for this purpose. The first scenario investigated driver reactions to system limitations during cornering. The results show that none of the subjects could avoid leaving their lane and moving into the oncoming lane and, therefore, could not control the situation safely. Due to partial automation, we could also identify a new part of the reaction time, the hands-on time, which leads to increased steering reaction times of 1.18 to 1.74 s. The second scenario examined driver responses to phantom braking caused by AEBS. We found that 25 of the 32 subjects could not override the phantom braking by pressing the accelerator pedal, although 16 subjects were informed about the system analog to the actual vehicle manuals. Overall, the study suggests that the current legal perspective on vehicle control and the expected driver reaction behavior for accident avoidance should be reconsidered.

This Is Your Brain on Autopilot 2.0: The Influence of Practice on Driver Workload and Engagement During On-Road, Partially Automated Driving.

This on-road study employed behavioral and neurophysiological measurement techniques to assess the influence of six weeks of practice driving a Level 2 partially automated vehicle on driver workload and engagement. Level 2 partial automation requires a driver to maintain supervisory control of the vehicle to detect "edge cases" that the automation is not equipped to handle. There is mixed evidence regarding whether drivers can do so effectively. There is also an open question regarding how practice and familiarity with automation influence driver cognitive states over time. Behavioral and neurophysiological measures of driver workload and visual engagement were recorded from 30 participants at two testing sessions-with a six-week familiarization period in-between. At both testing sessions, participants drove a vehicle with partial automation engaged (Level 2) and not engaged (Level 0) on two interstate highways while reaction times to the detection response task (DRT) and neurophysiological (EEG) metrics of frontal theta and parietal alpha were recorded. DRT results demonstrated that partially automated driving placed more cognitive load on drivers than manual driving and six weeks of practice decreased driver workload-though only when the driving environment was relatively simple. EEG metrics of frontal theta and parietal alpha showed null effects of partial automation. Driver workload was influenced by level of automation, specific highway characteristics, and by practice over time, but only on a behavioral level and not on a neural level. These findings expand our understanding of the influence of practice on driver cognitive states under Level 2 partial automation.

Earcons to reduce mode confusions in partially automated vehicles: Development and application of an evaluation method

Partially automated vehicles have systems that can ensure lateral and longitudinal control through adaptive cruise control and lane centering assist, meaning that there are three possible levels (modes) of automation: manual driving, automated longitudinal control, and automated lateral and longitudinal control. Confusions can occur when drivers fail to perceive or comprehend the automation mode indicated by the interface. Earcons reflecting the hierarchy of these automation modes through the use of pitch, rhythm, and number of note variations could improve mode awareness. The goal of the present study was therefore to develop and test a method for assessing the efficiency of earcons as indicators of automation mode, based on a mode awareness model. This method consisted in checking during a controlled laboratory situation, that the earcons were perceived correctly (Experiment 1), comprehended in isolation (Experiment 2), and comprehended during the performance of a parallel visual task that mimicked the visual demand of driving with partial automation and having to react to changes of visual icons (Experiment 3). This method was applied to earcons indicating the hierarchy of automation modes. The results of these experiments indicated that the earcons were efficiently perceived. They were comprehended in isolation and had a provoked a small decrement to a visual task. The parameters used here to create earcons for the purpose of reducing mode confusions were therefore validated to future simulator studies. The method for assessing them could be applied elsewhere, to ensure that other auditory signals adequately contribute to all dimensions of mode awareness before being integrated to a vehicle.

Keeping in the lane! Investigating drivers’ performance handling silent vs. alerted lateral control failures in monotonous partially automated driving

This study investigated how drivers can manage the take-over when silent or alerted failure of automated lateral control occurs after monotonous hands-off driving with partial automation. Twenty-two drivers with varying levels of prior ADAS experience participated in the driving simulator experiment. The failures were injected into the driving scenario on curved road segments, accompanied by either a visual-auditory alert or no change in HMI. Results indicated that drivers could rarely maintain lane-keeping when automated steering was disabled silently, but most drivers safely managed the alerted failure situation within the ego-lane. The silent failure yielded significantly longer take-over time and generally worse lateral control quality. In contrast, poor longitudinal control performance was observed in alerted conditions due to more brake usage. An expert-based controllability assessment method was introduced to this study. The silent lateral failure situation during monotonous hands-off driving was rated as uncontrollable, while the alerted situation was basically controllable. Participants showed their preferences for the TORs, and the importance of conveying TOR reasons was also demonstrated. Relevance to industryThe results and implications of this study provided insights into the design and development of automated driving systems to prevent critical consequences. The comprehensive method of controllability assessment can benefit the automated driving system evaluation.

Автоматизация выделения границ медицинских участков на примере Санкт-Петербурга

The article describes an approach to the desktop geographic information system application to optimization and (partial) automation of the spatial and non-spatial medical data processing and analysis when estimating the grid of medical lots in an area of medical service, and when forming (considering current morbidity and demographic indicators) updated boundaries of medical lots. Software tools of the QGIS platform were used to process spatial data (QGIS itself and its extension modules, including the GeoMedic module designed with the participation of the authors for medical statistics data geocoding). All medical statistics data used in the study were previously depersonalized in accordance with the requirements of Russian laws. A methodology prototype is elaborated for preparation, analysis, and processing of the data on patients (potential patients) number in the medical service area, together with spatial data (geographical map) of this area. The data of the St. Petersburg city tuberculosis service were used for experimental purposes in the study. Specialists of the St. Petersburg Research Institute of Phthisiopulmonology are involved in the work. Developed methodological toolset prototype ensures partially automated processing of initial data with the formation of boundaries of the medical lots in the analyzed area of medical service. Formation of the boundaries is carried out taking into account the normative number of patients, which must be ensured for the formed lots. The study was performed on the city scale, on the example of St. Petersburg (Russia). Data analysis was performed for the Moskovsky, Petrogradsky, Primorsky and Nevsky Administrative Districts of St. Petersburg. A grid of medical lot boundaries was formed and mapped in the results of the performed analysis. The methodology prototype proposed in the result of the conducted study can be applied in the analysis of existing schemas of medical lots and their updating.

Minimising workers’ workload in partially automated assembly lines with human-robot collaboration

The benefits of automation in human-operated assembly lines are well known. Despite the rapid progress made in the field, complete automation is not always possible. Often robots cannot replace workers’ flexibility, and sometimes there is no economically viable option for automation. A combined production system with both manual and automated parts could be the best configuration in such circumstances. Even with partial automation the number of workers in the line could be decreased, and workers’ safety and health conditions could be dramatically improved.In this paper, MILP models are suggested to compare the total workload of workers when different numbers of robots are applied in an assembly line with partial automation and the minimum possible number of workers. The proposed MILP model is implemented in the AIMMS modelling environment and solved using CPLEX.The presented models can help evaluate the impact of adding robots to the line on total workers’ workload and may assist decision makers in finding the best possible line configuration.

Perceived safety and trust in SAE Level 2 partially automated cars: Results from an online questionnaire.

The present online study surveyed drivers of SAE Level 2 partially automated cars on automation use and attitudes towards automation. Respondents reported high levels of trust in their partially automated cars to maintain speed and distance to the car ahead (M = 4.41), and to feel safe most of the time (M = 4.22) on a scale from 1 to 5. Respondents indicated to always know when the car is in partially automated driving mode (M = 4.42), and to monitor the performance of their car most of the time (M = 4.34). A low rating was obtained for engaging in other activities while driving the partially automated car (M = 2.27). Partial automation did, however, increase reported engagement in secondary tasks that are already performed during manual driving (i.e., the proportion of respondents reporting to observe the landscape, use the phone for texting, navigation, music selection and calls, and eat during partially automated driving was higher in comparison to manual driving). Unsafe behaviour was rare with 1% of respondents indicating to rarely monitor the road, and another 1% to sleep during partially automated driving. Structural equation modeling revealed a strong, positive relationship between perceived safety and trust (β = 0.69, p = 0.001). Performance expectancy had the strongest effects on automation use, followed by driver engagement, trust, and non-driving related task engagement. Perceived safety interacted with automation use through trust. We recommend future research to evaluate the development of perceived safety and trust in time, and revisit the influence of driver engagement and non-driving related task engagement, which emerged as new constructs related to trust in partial automation.

Investigating the decision to travel more in a partially automated electric vehicle

Abstract Partially automated battery electric vehicles (BEVs) are already being sold to and used by consumers. Estimates indicate that as of the end of 2019, there were over 1.2 million Partially Automated Tesla Vehicles—the subject of this study—on the roads globally. Despite this, little research has been done to understand how partially automated vehicles may be changing travel behaviour. In this study we conduct qualitative interviews with 35 owners of Tesla BEVs with Autopilot. The focus was to determine whether partially automated BEVs could cause or are causing an increase in travel. Results show that partial automation and electrification leads to interviewees driving more and choosing to drive rather than fly. These changes are due to increased comfort and reduced stress due to the partial automation system, and because of the lower running costs of a BEV. The results show how partially automated BEVs could increase vehicle miles travelled.

No Difference in Arousal or Cognitive Demands Between Manual and Partially Automated Driving: A Multi-Method On-Road Study.

IntroductionPartial driving automation is not always reliable and requires that drivers maintain readiness to take over control and manually operate the vehicle. Little is known about differences in drivers’ arousal and cognitive demands under partial automation and how it may make it difficult for drivers to transition from automated to manual modes. This research examined whether there are differences in drivers’ arousal and cognitive demands during manual versus partial automation driving.MethodWe compared arousal (using heart rate) and cognitive demands (using the root mean square of successive differences in normal heartbeats; RMSSD, and Detection Response Task; DRT) while 39 younger (M = 28.82 years) and 32 late-middle-aged (M = 52.72 years) participants drove four partially automated vehicles (Cadillac, Nissan Rogue, Tesla, and Volvo) on interstate highways. If compared to manual driving, drivers’ arousal and cognitive demands were different under partial automation, then corresponding differences in heart rate, RMSSD, and DRT would be expected. Alternatively, if drivers’ arousal and cognitive demands were similar in manual and partially automated driving, no difference in the two driving modes would be expected.ResultsResults suggest no significant differences in heart rate, RMSSD, or DRT reaction time performance between manual and partially automated modes of driving for either younger or late-middle-aged adults across the four test vehicles. A Bayes Factor analysis suggested that heart rate, RMSSD, and DRT data showed extreme evidence in favor of the null hypothesis.ConclusionThis novel study conducted on real roads with a representative sample provides important evidence of no difference in arousal and cognitive demands. Younger and late-middle-aged motorists who are new to partial automation are able to maintain arousal and cognitive demands comparable to manual driving while using the partially automated technology. Drivers who are more experienced with partially automated technology may respond differently than those with limited prior experience.

A fallback mechanism or a commander? A discussion about the role and skill needs of future drivers within partially automated vehicles

Vehicle automation promises to improve safety. However, it also raises concerns about human performance. Many current research and policy efforts address automation effects on operational levels of driving control. However, the driving task encompasses more than just operational control. So, in this paper, we ask whether a more holistic view of the driving task should be considered. Would such a framework imply that the human role is more than just a “fallback” mechanism for automation systems, especially for partial automation systems? If the human role were conceptualized more as a “commander role” (than an “operator”), would it enable system-level performance to be optimized more effectively? If so, would this lead to changes in design of the vehicle and interface—as well as in skills needed by human users? New skills like strategic planning, delegation of control responsibility, tactical avoidance, and effective mediation of hazardous situations could become important. A more holistic approach for integrating automation may facilitate role change for human users while also improving safety.

Driver Logo Sign Detection and Hazard Responses during Partially Automated Driving

This study investigates the presentation of service logo information under partially automated driving. Drivers completed simulated drives with partial automation during which they had to detect target logo signs and react to hazards by taking over vehicle control when needed. Driver performance was measured in terms of sign detection rate, crash rate, and hazard response time. A number of factors, including sign information source, sign information load, and driver age group, were investigated. In general, our findings support the delivery of service logo information via in-vehicle display under partially automated driving, especially when the in-vehicle display occurred simultaneously with the on-road signage. Under this presentation condition, drivers were most accurate in detecting target logo signs, and showed little impairment from processing sign information as a secondary task when negotiating a hazard. Implications of the findings and future directions were discussed.

User expectations of partial driving automation capabilities and their effect on information design preferences in the vehicle

Partially automated vehicles present interface design challenges in ensuring the driver remains alert should the vehicle need to hand back control at short notice, but without exposing the driver to cognitive overload. To date, little is known about driver expectations of partial driving automation and whether this affects the information they require inside the vehicle. Twenty-five participants were presented with five partially automated driving events in a driving simulator. After each event, a semi-structured interview was conducted. The interview data was coded and analysed using grounded theory. From the results, two groupings of driver expectations were identified: High Information Preference (HIP) and Low Information Preference (LIP) drivers; between these two groups the information preferences differed. LIP drivers did not want detailed information about the vehicle presented to them, but the definition of partial automation means that this kind of information is required for safe use. Hence, the results suggest careful thought as to how information is presented to them is required in order for LIP drivers to safely using partial driving automation. Conversely, HIP drivers wanted detailed information about the system's status and driving and were found to be more willing to work with the partial automation and its current limitations. It was evident that the drivers' expectations of the partial automation capability differed, and this affected their information preferences. Hence this study suggests that HMI designers must account for these differing expectations and preferences to create a safe, usable system that works for everyone.

The Effect of Partial Automation on Driver Attention: A Naturalistic Driving Study.

This naturalistic driving study investigated how drivers deploy visual attention in a partially automated vehicle. Vehicle automation is rapidly increasing across vehicle fleets. This increase in automation will likely have both positive and negative consequences as drivers learn to use the new technology. Research is needed to understand how drivers interact with partially automated vehicle systems and what impact new technology has on driver attention. Ten participants drove a Tesla Model S for 1 week during their daily commute on a stretch of busy interstate. Drivers were instructed to use Autopilot, a system that provides both lateral and longitudinal control, as much as they felt comfortable while driving on the interstate. Driver-facing video data were recorded and manually reduced to examine glance behavior. Drivers primarily allocated their visual attention between the forward roadway (74% of glance time) and the instrument panel (13%). With partial automation engaged, drivers made longer single glances and had longer maximum total-eyes-off-road time (TEORT) associated with a glance cluster. These results provide a window into the nature of visual attention while driving with partial vehicle automation. The results suggest that drivers may be more willing to execute long, "outlier" glances and clusters of glances to off-road locations with partial automation. The findings highlight several important human factors considerations for partially automated vehicles.

From partial and high automation to manual driving: Relationship between non-driving related tasks, drowsiness and take-over performance

BackgroundUntil the level of full vehicle automation is reached, users of vehicle automation systems will be required to take over manual control of the vehicle occasionally and stay fallback-ready to some extent during the drive. Both, drowsiness caused by inactivity and the engagement in distracting non-driving related tasks (NDRTs) such as entertainment or office work have been suggested to impair the driver’s ability to safely handle these transitions of control. Thus, it is an open question whether engagement in NDRTs will impair or improve take-over performance. MethodIn a motion-based driving simulator, 64 participants completed an automated drive that lasted either one or two hours using either a partially or highly automated driving system. In the partially automated driving condition, a warning was issued after several seconds when drivers took both hands off the steering wheel, while the highly automated driving system allowed hands-off driving permanently. Drivers were allowed to bring along their smartphones and to use them during the drive. They engaged in a wide variety of NDRTs such as reading or using social media. At the end of the session, drivers had to react to a sudden lead vehicle braking event. In the partial automation condition, there was no take-over request (TOR) to notify the drivers of the braking vehicle, while in the highly automated condition, the situation happened right after the drivers had deactivated the automation in response to a TOR. The lead time of the TOR was set at 8 s. Driver’s level of drowsiness, workload (visual, mental and motoric) from carrying out the NDRT and motivational appeal of the NDRT right before the control transition were video-coded and used to predict the outcome of the braking event (i.e., reaction and system deactivation times, minimal Time-to-collision (TTC) and self-reported criticality) with a multiple regression approach. ResultsIn the partial automation condition, reaction times to the braking vehicle and situation criticality as measured by the minimum TTC could be well predicted. Main predictors for increased reaction time were drowsiness and motivational appeal of the NDRT. However, visual and mental demand associated with NDRTs did decrease reaction time, suggesting that the NDRT helped the drivers to maintain alertness during the partially automated drive. Accordingly, drowsiness and motivational appeal of the NDRT increased situation criticality, while cognitive load due to the NDRT decreased it. In the highly automated condition, however, it was not possible to predict system deactivation time (in reaction to the TOR), brake reaction time to the braking vehicle and situation criticality by observed drowsiness and NDRT engagement. DiscussionThe results suggest a relationship between the driver’s drowsiness and NDRT engagement in partial automation but not in highly automated driving. Several explanations for this finding are discussed. It could be possible that the lead time of 8 s might have given the drivers enough time to complete the driver state transition process from executing NDRTs to manual driving, putting them in a position to be able to cope with the driving event, while this was not possible in the partial automation condition. Methodological issues that might have led to a non-detection of an effect of drowsiness or NDRT engagement in the highly automated driving condition, such as the sample size and sensitivity of the observer ratings, are also discussed.

Performance of an Additional Task During Level 2 Automated Driving: An On-Road Study Comparing Drivers With and Without Experience With Partial Automation.

To investigate the influence of prior experience with Level 2 automation on additional task performance during manual and Level 2 partially automated driving. Level 2 automation is now on the market, but its effects on driver behavior remain unclear. Based on previous studies, we could expect an increase in drivers' engagement in secondary tasks during Level 2 automated driving, but it is yet unknown how drivers will integrate all the ongoing demands in such situations. Twenty-one drivers (12 without, 9 with Level 2 automation experience) drove on a highway manually and with Level 2 automation (exemplified by Volvo Pilot Assist generation 2; PA2) while performing an additional task. In half of the conditions, the task could be interrupted (self-paced), and in the other half, it could not (system-paced). Drivers' visual attention, additional task performance, and other compensatory strategies were analyzed. Driving with PA2 led to decreased scores in the additional task and more visual attention to the dashboard. In the self-paced condition, all drivers looked more to the task and perceived a lower mental demand. The drivers experienced with PA2 used the system and the task more than the novice group and performed more overtakings. The additional task interfered more with Level 2 automation than with manual driving. The drivers, particularly the automation novice drivers, used some compensatory strategies. Automation designers need to consider these potential effects in the development of future automated systems.

An interview study exploring Tesla drivers' behavioural adaptation

Partially automated vehicles (PAVs) have been used in real-world environments for several years since the emergence of autonomous driving. It is important to understand the effects of partial automation systems (PAS) on the understanding of drivers and their behaviour during the first months of use. In order to adapt to new vehicle technology, drivers usually exhibit specific behaviours in this stage that are not intended by the developers, namely behavioural adaptation. The present study investigated the behavioural adaptations by early PAV adopters after short-term usage. A semi-structured interview was conducted among 20 Tesla drivers who had relatively high experience (one to five months) with Autopilot, and the interviews were synthesized to understand their behavioural adaptation, mental models, and trust during the period of use. The results showed that PAV drivers had a very positive attitude towards the PAS and drivers universally engaged in secondary tasks during automated driving. They also learned from their experiences to identify relatively safe usage conditions and they employed a safety margin to avoid exposure to excessively risky situations.

The impact of an in-vehicle display on glance distribution in partially automated driving in an on-road experiment

One of the major challenges of designing an HMI for partially automated vehicles is the trade-off between a sufficient level of system information and avoidance of distracting the driver. This study aimed to investigate drivers’ glance behavior as an indicator of distraction when vehicle guidance is partially automated. Therefore, an on-road experiment was conducted comparing two versions of an in-vehicle display (during partially automated driving) and no display (during manual driving) on a heavy congested highway segment. The distribution of drivers’ total glance durations on the HMI showed that visual attention was shifted away from monitoring the central road scene towards looking at the in-vehicle display to a considerable extent. However, an analysis of the distribution of single glance durations supports the view that using partial automation and a respective HMI does not lead to a critical increase in distraction. Driving with a simplified version of the HMI had the potential to reduce glance duration on and thus potential distraction of the in-vehicle display.

GlidePath: Eco-Friendly Automated Approach and Departure at Signalized Intersections

Recently, there has been significant research on environment-focused connected vehicle (CV) applications that involve determining optimal speed profiles for vehicles traveling through signalized intersections and conveying this information to drivers via driver-vehicle interfaces (DVI's). However, findings from previous studies indicate that drivers may not be able to precisely follow the recommended speed profiles, resulting in degraded effectiveness of the applications. Moreover, the DVI could be distracting, which may compromise safety. As an alternative, partial automation can play an important role in ensuring that the benefits of these CV applications are fully realized. In this study, a partially automated vehicle system with an eco-approach and departure feature (called the GlidePath Prototype), which can receive dedicated short range communication message sets from the intersection and automatically follow recommended speed profiles, was developed, demonstrated, and evaluated. The results revealed that compared to manually following the recommended speed profiles, the GlidePath Prototype reduced fuel consumption by 17% on average. In some cases, the fuel savings are greater than 40% while the travel time is shortened by up to 64%. Furthermore, the system potentially improved the driving comfort since it would smooth out the speed profiles.