SAE Level Research Articles

Drivers’ usage of driving automation systems in different contexts: A survey in China, Germany, Spain and USA

AbstractThe purpose of the study was to investigate how drivers use assisted and automated driving systems (DAS), more specifically their usage of SAE Level 1 and Level 2 systems, in different situations. An online survey was distributed to 2500 respondents in China, Germany, Spain, and the USA. The final dataset consisted of 549 respondents, all non‐professional drivers, with access to a minimum of a Level 1 system. A subset, 159 in total, had access also to a Level 2 DAS. The survey included questions on the attitude towards, access to, and usage of Level 1 and Level 2 systems in nine different situations respectively. The data was analysed on an individual and a national level. A cluster analysis showed two main groups: frequent and non‐frequent users. On an individual level, the reported usage of Level 1 and Level 2 DAS respectively differed depending on traffic situation, weather and daylight conditions and driver state. Reports by respondents with access to both Level 1 and Level 2 systems did not reveal any statistically significant differences in usage between situations. The Spanish sample was the only one that showed a consistently different usage pattern compared to samples from China, Germany, and the USA.

Effects of brand and brand trust on initial trust in fully automated driving system.

Before Automated Driving Systems (ADS) with full driving automation (SAE Level 5) are placed into practical use, the issue of calibrating drivers' initial trust in Level 5 ADS to an appropriate degree to avoid inappropriate disuse or improper use should be resolved. This study aimed to identify the factors that affected drivers' initial trust in Level 5 ADS. We conducted two online surveys. Of these, one explored the effects of automobile brands and drivers' trust in automobile brands on drivers' initial trust in Level 5 ADS using a Structural Equation Model (SEM). The other identified drivers' cognitive structures regarding automobile brands using the Free Word Association Test (FWAT) and summarized the characteristics that resulted in higher initial trust among drivers in Level 5 ADS. The results showed that drivers' trust in automobile brands positively impacted their initial trust in Level 5 ADS, which showed invariance across gender and age. In addition, the degree of drivers' initial trust in Level 5 ADS was significantly different across different automobile brands. Furthermore, for automobile brands with higher trust in automobile brands and Level 5 ADS, drivers' cognitive structures were richer and varied, which included particular characteristics. These findings suggest the necessity of considering the influence of automobile brands on calibrating drivers' initial trust in driving automation.

Is Users’ Trust during Automated Driving Different When Using an Ambient Light HMI, Compared to an Auditory HMI?

The aim of this study was to compare the success of two different Human Machine Interfaces (HMIs) in attracting drivers’ attention when they were engaged in a Non-Driving-Related Task (NDRT) during SAE Level 3 driving. We also assessed the value of each on drivers’ perceived safety and trust. A driving simulator experiment was used to investigate drivers’ response to a non-safety-critical transition of control and five cut-in events (one hard; deceleration of 2.4 m/s2, and 4 subtle; deceleration of ~1.16 m/s2) over the course of the automated drive. The experiment used two types of HMI to trigger a takeover request (TOR): one Light-band display that flashed whenever the drivers needed to takeover control; and one auditory warning. Results showed that drivers’ levels of trust in automation were similar for both HMI conditions, in all scenarios, except during a hard cut-in event. Regarding the HMI’s capabilities to support a takeover process, the study found no differences in drivers’ takeover performance or overall gaze distribution. However, with the Light-band HMI, drivers were more likely to focus their attention to the road centre first after a takeover request. Although a high proportion of glances towards the dashboard of the vehicle was seen for both HMIs during the takeover process, the value of these ambient lighting signals for conveying automation status and takeover messages may be useful to help drivers direct their visual attention to the most suitable area after a takeover, such as the forward roadway.

How certain are we that our automated driving system is safe?

Objective Regulations are currently being drafted by the European Commission for the safe introduction of automated driving systems (ADSs) with conditional or higher automation (SAE level 3 and above). One of the main challenges for complying with the drafted regulations is proving that the residual risk of an ADS is lower than the existing state of the art without the ADS and that the current safety state of European roads is not compromised. Therefore, much research has been conducted to estimate the safety risk of ADS. One proposed method for estimating the risk is data-driven, scenario-based assessment, where tests are partially automatically generated based on recorded traffic data. Although this is a promising method, uncertainties in the estimated risk arise from, among others, the limited number of tests that are conducted and the limited data that have been used to generate the tests. This work addresses the following question: “Given the limitations of the data and the number of tests, what is the uncertainty of the estimated safety risk of the ADS?” Methods To compute the safety risk, parameterized test scenarios are based on large-scale collections of road scenarios that are stored in a scenario database. The exposure of the scenarios and the parameter distributions are estimated using the data as well as confidence bounds of these estimates. Next, virtual simulations are conducted of the scenarios for a variety of parameter values. Using a probabilistic framework, all results are combined to estimate the residual risk as well as the uncertainty of this estimation. Results The results are used to provide confidence bounds on the calculated fatality rate in case an ADS is implemented in the vehicle. For example, using the proposed probabilistic framework, it is possible to claim with 95% certainty that the fatality rate is less than fatalities per hour of driving. The proposed method is illustrated with a case study in which the risk and its uncertainty are quantified for a longitudinal controller in 3 different types of scenarios. The case study code is publicly available. Conclusions If results show that the uncertainty is too high, the proposed method allows answering questions like “How much more data do we need?” or “How many more (virtual) simulations must be conducted?” Therefore, the method can be used to set requirements on the amount of data and the number of (virtual) simulations. For a reliable risk estimate, though, much more data are needed than those used in the case study. Furthermore, because the method relies on (virtual) simulations, the reliability of the result depends on the validity of the models used in the simulations. The presented case study illustrates that the proposed method is able to quantify the uncertainty of the estimated safety risk of an ADS. Future work involves incorporating the proposed method into the type approval framework for future ADSs of SAE levels 3, 4, and 5, as proposed in the upcoming European Union implementing regulation for ADS.

Implications of the positive risk balance on the development of automated driving

Objectives Automated driving (AD) from SAE level 3 onwards represents a paradigm change from human driver controlling the vehicle to a technical system controlling it. In this light, different regulatory bodies (European Commission, Germany, etc.) have defined guidelines for the operation of such a system. One core principle of these guidelines is that the automated operation needs to be at least as safe as human driving—often referred to as the “positive risk balance.” However, these guidelines are general and do not provide details on what this means in a practical sense. This article discusses a method to demonstrate how positive risk balance can be addressed in practice. Methods Starting from a detailed analysis of corresponding guidelines and a literature review of possible risk assessment frameworks, a comprehensive approach has been developed to consider ethical requirements for the development of AD. This approach covers different development stages. The PrOACT-URL (Problems, Objectives, Alternatives, Consequences, Trade-offs, Uncertainty, Risk attitudes, and Linked decisions) approach was chosen for reporting of the work. Results The article will present the approach developed by BMW to ensure that a positive risk balance is achieved for an AD system. The approach is presented per development stage (concept phase, AD development phase, verification and validation phase, post–start of production phase). In the concept phase, the scope is to define how good a human driver is and how good an AD needs to be. In the AD development phase, first the relevant system requirements need to be derived. Monte Carlo experiments in combination with Bayesian networks are applied. The fulfillment of these requirements is checked in the verification phase through simulations and test track and real-world tests. For validation of the risk balance, the impact of AD in terms of traffic safety is derived by means of simulation. In the post–start of production phase, field observation is used. Conclusion The safety of AD is paramount when it comes to its operation and ensuring trust in this technology. The described approach contributes directly to building this trust by considering the principle of a positive risk balance throughout the development in addition to existing safety standards for advance driver assistance systems, such as ISO 26262, ISO21434 or ISO 21488.

Using pupillometry and gaze-based metrics for understanding drivers’ mental workload during automated driving

This Horizon2020-funded driving simulator-based study on automated driving investigated the effect of different car-following scenarios, and takeover situations, on drivers’ mental workload, as measured by eye tracking-based metrics of pupil diameter and self-reported workload ratings. This study incorporated a mixed design format, with 16 drivers recruited for the SAE Level 2 (L2; SAE International, 2021) automation group, who were asked to monitor the driving and road environment during automation, and 16 drivers in the Level 3 (L3) automation group, who engaged in a non-driving related task (NDRT; Arrows task) during automation. Drivers in each group undertook two experimental drives, lasting about 18 min each. To manipulate perceived workload, difficulty of the driving task was controlled by incorporating a lead vehicle which maintained either a Short (0.5 s) or Long (1.5 s) Time Headway (THW) condition during automated car-following (ACF). Each ACF session was followed by a subsequent request to takeover, which happened either in the presence or absence of a lead vehicle. Results from standard deviation of pupil diameter values indicated that drivers’ mental workload levels fluctuated significantly more when monitoring the drive during L2 ACF, compared to manual car-following (MCF). Additionally, we found that drivers’ mental workload, as indicated by their mean pupil diameter, increased steeply around takeovers, and was further exacerbated by the presence of a lead vehicle during the takeovers, especially in the Short THW condition, for both groups. Pupil diameter was found to be sensitive to subtle variations in mental workload, and closely resembled the trend seen in self-reported workload ratings. Further research is warranted to assess the feasibility of using eye-tracking-based metrics along with other physiological sensors, especially in real-world settings, to understand whether they can be used as real-time indicators of drivers’ mental workload, in future driver state monitoring systems.

The Space Age

Germany has become the first country in the world to offer automated valet parking to the public – made possible via SAE Level 4 legislation, connectivity services and hardware from Bosch and Mercedes

Normalizing crash risk of partially automated vehicles under sparse data

The safety of increasingly automated vehicles is of great concern to regulators, yet crash rates are generally reported by manufacturers using proprietary metrics with limited source data. Without consistent definitions of crashes and exposure, automated vehicle crash rates cannot be meaningfully compared with baseline datasets. The objective of this study was to establish methods to normalize automated vehicle crash rates using one manufacturer’s crash reports as a case study. The manufacturer’s quarterly crash rates for vehicles using SAE Level 1 and Level 2 automation were compared. Road type was controlled for using data from a naturalistic driving study with the same model vehicles, while driver age was controlled for using demographic ownership surveys. Although Level 2 vehicles were claimed to have a 43% lower crash rate than Level 1 vehicles, their improvement was only 10% after controlling for different rates of freeway driving. Direct comparison with general public driving was impossible due to unclear crash severity thresholds in the manufacturer’s reports, but analysis showed that controlling for driver age would increase reported crash rates by 11%. These results establish the need for detailed crash data, crash definitions, and exposure and demographic data in order to accurately assess automated vehicle safety.

Social Perception of Autonomous Vehicles

The technology of autonomous vehicles is gaining more and more emphasis these days. In the near future the technological developments will make it possible for vehicles to travel on the roads without human intervention. However, downstream users have differing views on this new mode of transport. The aim of our research was to explore the opinions of different social generation groups and traffic groups about fully autonomous self-driving (SAE Level 5). In our research, we conducted an online self-report questionnaire survey. The questionnaire was completed by 223 people. The results were analyzed from several perspectives. The results showed that opinions and expectations in the field of autonomous vehicles differed by generation group, gender and primary mode of transport.

Pilot study: Effect of roles and responsibility training on driver's use of adaptive cruise control between younger and older adults

With the development of driver support systems (SAE Levels 1 – 2), drivers must take on new monitoring and supervision tasks in additional to manual driving. Training is necessary to clarify drivers' new roles and promote safe usage and trust in these systems. Providing training for lower-levels of automation may also benefit drivers’ acceptance of future Fully Automated Vehicles (FAVs, SAE Level 5). However, younger and older drivers differ in training preferences (e.g., owner's manual vs on-road trial and error) and hold different attitudes towards automation. This study investigates the effects of additional training on drivers' roles and responsibilities when using Adaptive Cruise Control (ACC, SAE Level 1) for younger and older drivers. Thirty-nine adults (20 younger + 19 older) were trained on one of two ACC training protocols: basic (system functionality, operational procedures, and limitations) and comprehensive (basic training + ACC background and roles of responsibilities). Participants’ situational trust and ACC usage was evaluated before, during, and after experiencing an emergency event while using ACC in a driving simulator study. Results showed that the comprehensive training promoted drivers' situational trust in ACC, ACC usage, and the acceptance of FAvs Compared to younger drivers, older drivers used ACC less, reported less dynamic situational trust, higher levels of workload, and lower acceptance. Overall, comprehensive training resulted in older drivers behaving similarly to younger drivers. The comprehensive training also promoted the acceptance of FAVs for both younger and older drivers. In conclusion, training of drivers’ roles and responsibilities has an impact on drivers’ usage of ACC and may be particularly useful for older drivers.

Should my automated car drive as I do? Investigating speed preferences of drivengers in various driving conditions.

Studies investigating the question of how automated cars (ACs) should drive converge to show that a personalized automated driving-style, i.e., mimicking the driving-style of the human behind the wheel, has a positive influence on various aspects of his experience (e.g., comfort). However, few studies have investigated the fact that these benefits might vary with respect to driver-related variables, such as trust in ACs, and contextual variables of the driving activity, such as weather conditions. Additionally, the context of intermediate levels of automation, such as SAE level 3, remains largely unexplored. The objective of this study was to investigate these points. In a scenario-based experimental protocol, participants were exposed to written scenarios in which a character is driven by a SAE level 3 AC in different combinations of conditions (i.e., types of roads, weather conditions and traffic congestion levels). For each condition, participants were asked to indicate how fast they would prefer their AC to drive and how fast they would manually drive in the same situation. Through analyses of variance and equivalence tests, results showed a tendency for participants to overall prefer a slightly lower AC speed than their own. However, a linear regression analysis showed that while participants with the lowest levels of trust preferred an AC speed lower than theirs, those with the highest levels preferred an AC speed nearly identical to theirs. Overall, the results of this study suggest that it would be more beneficial to implement a personalization approach for the design of automated driving-styles rather than a one for all approach.

Adding visual contextual information to continuous sonification feedback about low-reliability situations in conditionally automated driving: A driving simulator study

In conditionally automated driving, control transitions where drivers are asked to take over manual control quickly are a known challenge as drivers are likely “out-of-the-loop” due to automated driving. We previously demonstrated a useful sonification feedback design to support drivers for takeover events by manipulating background music to convey real-time changes in the automation reliability level. The present research examines two types of auxiliary information—explanations about the potential sources for reduced reliability and recommended actions to prepare for a potential takeover event—to communicate event-specific context about the low-reliability situations indicated by the sonification feedback. In a driving simulator with SAE level 3 automation, forty participants were assigned to one of four interface conditions: (i) sonification feedback only, (ii) sonification feedback and visual auxiliary information providing explanations, (iii) sonification feedback and visual auxiliary information providing action recommendations, or (iv) sonification feedback and both explanatory and action-recommending auxiliary information. Participants experienced six low-reliability events that belonged to three types of situations that may eventually lead to a takeover event: operational problems, system limitations, and unexpected events. Two events became system-requested takeover events. Our results showed that auxiliary information impacted visual information processing but not the takeover decisions or response times to takeover requests. The situation type—characterized by the availability of cues and presence of hazards in the environment—also played a significant role in both participants’ takeover decisions and visual information processing behaviors. These findings have implications for designing effective feedback for not only Level 3 automation, which we studied, but also for Level 4 dual-mode vehicles that could allow transitions to manual driving when exiting their operational design domains and for Level 2 automation, where drivers may need more support to monitor the road effectively and make more timely takeover decisions.

Infrastructure requirements for the safe operation of automated vehicles: Opinions from experts and stakeholders

The need to future-proof road transport networks is becoming increasingly urgent in order to take full advantage of automated vehicles (AVs). It is now vital to understand the basic road infrastructure requirements of AVs in order to assess the readiness of the existing road network and prepare the roads for the safe operation of these vehicles. However, current literature on this subject is limited. As such this research seeks to understand the desired infrastructure-related requirements of highly automated vehicles (SAE Level 4) for safe operation based on a survey with experts and stakeholders. On the basis of 168 expert responses from 29 countries, this study presents stakeholders’ views on: (1) deployment paths of Level 4 automated driving, (2) the concept of road certification for automated driving, (3) basic road infrastructure elements for the safe operation of automated driving, and (4) factors affecting safe operations of Level 4 automated driving. The findings show that different types of stakeholders (e.g. academics, infrastructure owners and operators, and vehicle and information technology developers) have broadly similar views on most criteria requiring consideration in the early stages of automated driving implementation. However, there is no clear consensus on issues regarding operating constraints on road networks and some are in favour of waiting for the technology to mature or until it proves beneficial to overall community goals and then acting on the infrastructure needs.

Which not-at-fault crashes are unavoidable by using current active safety technology?

Previous research has focused on analysing crash reduction potential of active safety technology in at-fault passenger cars, but only a few studies have examined counterparties' possibilities to avoid collisions by using advanced driver assistance systems (ADAS). This study quantified the incidence of fatal not-at-fault passenger car crashes that current ADAS (up to SAE level 2) would be unable to avoid. We used data taken from in-depth investigated fatal crashes in which a passenger car was involved, that car having been first registered during the period 1 January 2010 to 31 December 2017 in Finland. The evaluation of unavoidable crashes consisted of two evaluation rounds. The preliminary evaluation round identified potential active safety systems that could have operated in the studied cases. In the following round, we made a five-level case-by-case analysis including a time headway analysis in order to evaluate the possibilities for crash avoidance. The crash data included 63 fatal crashes, of which five were excluded because the death was due to a sudden illness attack. The remaining 58 crashes were classified as follows: probably unavoidable (n = 51), avoidable (n = 3), and unclear (n = 4). The crash incidence of the unavoidable not-at-fault party crashes was 0.67–0.73 fatal crashes per billion kilometers and 14–15 fatal crashes per million registration years. The results indicate that current active safety systems may be able to prevent not-at-fault party fatal crashes only in a few cases and that the driver's role in road safety remains important despite the deployment of the active safety systems.

Understanding Intention to Use Conditionally Automated Vehicles in Thailand, Based on an Extended Technology Acceptance Model

Automated vehicles (AVs) provide several advantages in solving issues of road traffic; including enhanced safety, reduced greenhouse gas emissions, and reduced traffic congestion. As AVs are still relatively new developments in developing countries, AV adoption faces challenges from both technological and psychological issues. Therefore, our initial research focus is on identifying the factors that influence the intention to use conditionally automated vehicles (CAVs; SAE Level 3). An extended technology acceptance model (TAM), which includes Trust, Perceived Risks, and Environmental concerns, is proposed as the predictor model in this study. The 299 participants gathered through online surveys in Thailand were examined using the Structural Equation Model (SEM) technique. In this study, Trust was shown to be the strongest predictor of Intention, followed by Perceived Ease of Use, whereas Perceived Usefulness had no impact on intention to use the SAE Level 3. The results of this study will be able to guide the forming of future policies that aim at promoting the use of AVs and helping technology developers create systems to better meet the needs of users in developing nations.

Exploration into the Needs and Requirements of the Remote Driver When Teleoperating the 5G-Enabled Level 4 Automated Vehicle in the Real World-A Case Study of 5G Connected and Automated Logistics.

Connected and automated vehicles have the potential to deliver significant environmental, safety, economic and social benefits. The key advancement for automated vehicles with higher levels of automation (SAE Level 4 and over) is fail-operational. One possible solution for the failsafe mode of automated vehicles is a 5G-enabled teleoperation system controlled by remote drivers. However, knowledge is missing regarding understanding of the human-machine interaction in teleoperation from the perspective of remote drivers. To address this research gap, this study qualitatively investigated the acceptance, attitudes, needs and requirements of remote drivers when teleoperating a 5G-enabled Level 4 automated vehicle (5G L4 AV) in the real world. The results showed that remote drivers are positive towards the 5G L4 AV. They would like to constantly monitor the driving when they are not controlling the vehicle remotely. Improving their field of vision for driving and enhancing the perception of physical motion feedback are the two key supports required by remote drivers in 5G L4 AVs. The knowledge gained in this study provides new insights into facilitating the design and development of safe, effective and user-friendly teleoperation systems in vehicle automation.

Do males and females prefer different non-driving related activities during automated driving?

We investigated gender differences in the willingness to use travel time for Non-Driving Related Tasks (NDRs) in conditionally automated vehicles (SAE Level 3) using a questionnaire study among 8,412 car-drivers in eight European countries (Germany, Italy, Netherlands, Poland, Romania, Spain, Sweden, and the United Kingdom), as part of the European L3Pilot project. Participants’ responses to what activities they would like to perform while the automated driving is activated were investigated. Results reveal that females were more willing to engage in NDRTs than males for nine out of 12 investigated activities. This study informs about gender differences in engagement with NDRTs which will help in policy development and future interior vehicle design to ensure that all potential users can benefit from AVs.

Deriving Environmental Risk Profiles for Autonomous Vehicles From Simulated Trips

The commercial adoption of Autonomous Vehicles (AVs) and the positive impact they are expected to have on traffic safety depends on appropriate insurance products due to the high potential losses. A significant proportion of these losses are expected to occur from the out-of-distribution risks which arise from situations outside the AV’s training experience. Traditional vehicle insurance products (for human-driven vehicles) rely on large data sets of drivers’ background and historical incidents. However, the lack of such datasets for AVs makes it imperative to exploit the ability to deploy AVs in simulated environments. In this paper, the data collected by deploying Autonomous Driving Systems (ADSs) in simulated environments is used to develop models to answer two questions: (1) <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">how risky a road section is for an AV to drive?</i> and (2) <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">how does the risk profile vary with different (SAE levels) of ADSs?</i> A simulation pipeline was built on the CARLA (Car Learning to Act): an open-source simulator for autonomous driving research). The environment was specified using parameters such as <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">weather, lighting, traffic density, traffic flow, no. of lanes</i> , etc. A metric - <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">risk factor</i> was defined as a combination of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">harsh accelerations/braking, inverse Time to Collision</i> , and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">inverse Time Headway</i> to capture the crashes and near crashes. To assess the difference between ADSs, two ADSs: OpenPilot (Level 2/3) and Pylot (Level 4) were implemented in the simulator. The results (from data and model predictions) show that the trends in the relation between the environment <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">features</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">risk factor</i> for an AV are similar to those observed for human drivers (e.g., risk increases with <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">traffic flow</i> ). The models also showed that <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">junctions</i> were a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">risk hot-spot</i> for both ADSs. The feature importance of the model revealed that the Level 2/3 ADS is more sensitive to <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">no. of lanes</i> and the Level 4 ADS is sensitive to <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">traffic flow</i> . Such differences in feature importance provide valuable insights into the risk characteristics of different ADSs. In the future, this base model will be extended to include other features (other than the environment), e.g., take over requests, and also address the deficiencies of the current simulation data in terms of insensitivity to <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">weather</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">lighting</i> .

Toward Fault-Tolerant Vehicle Motion Control for Over-Actuated Automated Vehicles: A Non-Linear Model Predictive Approach

Automated driving systems operated at SAE levels 4 and 5 require a far-reaching fault-tolerant design. To meet this need at the actuator level, we present an integrated vehicle motion control approach that is able to tolerate a wide range of different actuator degradations and failures as well as tire blowouts in vehicles featuring four wheel-individual steering, drive, and brake actuators. The approach, which is based on non-linear model predictive control (MPC), tracks a temporal sequence of reference poses. Fault tolerance is achieved by reconfiguration of the MPC’s constraints, weights, and prediction model, which consists of a double-track vehicle and a brush tire model. The evaluation of the approach is based on two reference trajectories. The example of a simple single lane change trajectory in IPG CarMaker demonstrates the basic functionality of the approach. The example of a demanding decelerated single lane change trajectory shows that the approach is subject to limitations when tolerating different degradations and failures. Still, the observed limitations can be explained by the interplay of the specific degradation or failure and the demanding nature of the trajectory. Therefore, the results indicate that the suitability of fault-tolerant vehicle motion control as part of a system-wide safety concept is strongly connected to the range of possible driving scenarios that an automated driving system can encounter.

Optimization-based Fault Mitigation for Safe Automated Driving

With increased developments and interest in cooperative driving and higher levels of automation (SAE level 3+), the need for safety systems that are capable to monitor system health and maintain safe operations in faulty scenarios is increasing. A variety of faults or failures could occur, and there exists a high variety of ways to respond to such events. Once a fault or failure is detected, there is a need to classify its severity and decide on appropriate and safe mitigating actions. To provide a solution to this mitigation challenge, in this paper a functional-safety architecture is proposed and an optimization-based mitigation algorithm is introduced. This algorithm uses nonlinear model predictive control (NMPC) to bring a vehicle, suffering from a severe fault, such as a power steering failure, to a safe-state. The internal model of the NMPC uses the information from the fault detection, isolation and identification to optimize the tracking performance of the controller, showcasing the need of the proposed architecture. Given a string of ACC vehicles, our results demonstrate a variety of tactical decision-making approaches that a fault-affected vehicle could employ to manage any faults. Furthermore, we show the potential for improving the safety of the affected vehicle as well as the effect of these approaches on the duration of the manoeuvre.