Abstract Software defect prediction using code metrics has been extensively researched over the past five decades. However, prediction using non-software metrics remains under-researched. Given that the root cause of software defectsis often attributed to human error, human factors theory might offer key forecasting metrics for actionable insights. This paper explores automated software defect prediction based on developers' coding habits. First, we proposea framework for deciding which metrics to use for prediction. Next, we comparethe performance of our metrics to the state-of-the-art code metrics. We then analyze the predictive importance of each metric. Our analysis of twenty-one critical infrastructure, large-scale, open-source software projects yielded the following results: 1) We have presented a human-error-based framework with metrics useful for defect prediction at the method level. 2) Models using our proposed metrics achieved a better average prediction performance than state-of-the-art code metrics and history measures. 3) The predictive importanceof all metrics distributes differently, with each of the novel metrics having abetter average importance than code and history metrics. 4) Our novel metrics dramatically enhance the explainability, practicality, and actionability of software defect prediction models, which significantly advances the field.We present a systematic approach to forecasting defect-prone software methodsvia a human error framework. This work empowers practitioners to act on predictions, empirically demonstrating how developer coding habits contribute todefects in software systems.

Automation is defined as the replacement of tasks previously handled by humans with machines [1]. One value of automated driving is that it enables humans to engage in non-driving tasks, thereby allowing for more efficient use of time. However, humans may suffer from boredom created by monotonous environment or drowsiness due to human characteristics in conditional driving automation, where the role of resuming DDTs is assigned to humans. Humans may have an intention of arriving at their destination faster. Humans may underestimate risks associated with drowsiness and may not form an intention to take a break. Humans may have their attention focused on non-driving tasks and fail to find an opportunity for taking a break. The automated driving system should not continue to vehicle control in a situation where humans, who are expected to act as a backup, feel drowsy. One technology that is currently lacking is a mechanism for humans and machines to communicate their intentions and form joint decisions for strategic, tactical, and operational levels. The mechanism is called mediation, meaning means of interaction strategies aimed to reach a joint intent and an adequate action. Against this background, the purpose of the study is to design human-machine mediation for joint intention formation in coping with drowsiness under driving automation. We constructed a state description of human-machine interaction based on relationship between environment, human intentions and behaviors, and machine intentions and behaviors. As shown in Fig. 1, the flow of interactions that make a joint intent and an adequate action for responding to driver drowsiness was described using Swimlane Diagrams [2]. The Karolinska Sleepiness Scale [3] was used to determine sleepiness levels, and opportunities for human-machine interaction were constructed according to sleepiness levels and levels of automation (LOA) [4]. In an era of increasing autonomy, technologies that support consensus-building between humans and machines are important, and the contribution of this study was to present a method for describing the state of human-machine interactions, using the use case of coping with drowsiness. We plan to evaluate the usability of the designed mediation in future experiments.

The utilization of gaze behavior for control has been studied as one of the hands-free control methods. With the recent spread of Head Mounted Display devices, it has become a vital issue to establish hands-free control methods. Previously proposed approaches using the eye gaze behavior have several problems to be solved for natural manipulation during content browsing. Therefore, in this study, considering the case where the gaze point leaves the decision area for a short period, we propose a method that enables accurate control prediction based on gaze behavior during content browsing by counting the cumulative gaze dwell time. Through experimentation, we confirmed that the proposed method would be effective for the accurate prediction of user intention. The limitation of our method and further issues to be addressed were also clarified.

Partial driving automation systems execute sustained lateral and longitudinal control, while humans are required to supervise the automated driving features. Similar with drivers using manual driving settings, those using automated driving systems may also experience drowsiness. However, existing systems that aim to detect driver drowsiness tend to be unreliable. This study applies a dual-control scheme in the partial driving automation context in which the driver must keep their hands on the vehicle’s steering wheel. This scheme executes a partial steering control when a vehicle lane departure is anticipated (because of inappropriate torque input), and then, activates a deceleration control if the driver does not properly perform the required action. To determine whether the driver is supervising the partial driving automation, the scheme attempts to create an opportunity for driver–automation interactions. Thus, the controller’s objectives are twofold: safety control and driver state identification. This study investigated the effectiveness of the scheme for identifying driver drowsiness and preventing lane departures using only vehicle information. Twenty drivers participated in a fixed-base driving simulator experiment in a sleep-inducing environment. While we observed cases in which the system could effectively bring the vehicle to a controlled stop, the timeliness and accuracy of the driver state identification remained as issues owing to indirect links between the drivers’ drowsiness level and controller activation. We conclude that although the dual-control scheme is a useful mechanism to avoid lane departures, the driver state identification needs to be improved to ensure timely and effective detection of driver drowsiness.

In the context of hands-on partial driving automation, drivers engage in a combination of lane centering system (LCS) and adaptive cruise control (ACC). In lane change scenarios, what should the system do, if it detects the approach of vehicle in the blind spot? Humans with powerful and intelligent machine sometimes suffer from such as automation surprises and distrust in automation. By contrast, haptic and continuous information would enable the driver to virtually touch their potentially hazardous environment. In this study, we propose the haptic protection with an adaptive level of authority that can be turned based on risk indicators, i.e., TTC and THW , in the blind spot scenarios. In haptic guidance systems, the artificial stiffness around the controller’s desired steering angle is used as the level of haptic authority to guide the driver to normative behavior. On the other hand, in the proposed protection system, the artificial stiffness around the driver’s desired steering angle is used as the level of haptic authority to warn the driver about their erroneous action. Thus, the definition of the level of haptic authority (LoHA) is different between the guidance-type and protection-type in terms of the objective. In a simulation with a parameter space of lane change scenarios, this study highlighted the combined impact of the haptic guidance and haptic protection on the driver bahavior. Results indicated that the use of the proposed haptic protection can enable the drivers to recognize their inappropriate actions.

This article focuses on explanation-based knowledge about system limitations (SLs) under conditional driving automation (society of automotive engineers level 3) and aims to reveal how this knowledge influences driver intervention. By illustrating the relationships between the driving environment, system, and mental model, knowledge in dynamic decision-making processing for responding to an issued request to intervene (RtI), occurrence of SL, concept of RtI, and scene(s) related to SL are determined by knowledge-based learning. Based on three concepts, the knowledge is examined at five levels: 1) no explanation, 2) occurrence of SL, 3) concept of RtI, 4) some typical scenes related to SL, and 5) all of the above. Data collection is conducted on a driving simulator, and 100 people with no experience of automated driving participated. The experimental results show that instructing drivers in typical situations contributes to a greater increase in the rate of successful intervention in car control from 55% to 95%. Furthermore, instructing them on the concept of RtI is conducive to a significant reduction in response time from 5.48 to 3.62 s in their first experience of RtI. It is also revealed that the knowledge-based learning effect dwindles but does not vanish even after drivers experience RtI a number of times. Compared to explaining all possible situations to a driver, introducing typical situations results in better take-over performances even in critical or unexplained scenarios. This article demonstrates the importance and necessity of this knowledge, especially the explanation of sample scenes related to SL, which contributes to drivers' take-over behavior.

Driving automation leads drivers to be disengaged by dynamic driving tasks, but the need for the driver to monitor traffic environment is existed in a partial automation. Monotonous road situations can produce drowsiness, resulting in poor driving performance. Conversation in automated vehicles is expected to mitigate fatigued driving. However, few studies have examined effects of conversation. This paper reports a driving simulator study describing impacts of passenger conversation on driver fatigue and driving performance regarding the relationship between a driver and a passenger. Participants were categorized into two groups: a passenger's friends (ConvR) and people who had not met the passenger before (ConvNR) to examine whether the acquaintance with the passenger affects driver behaviour, driving performance, fatigue and workload. The results showed that drivers in ConvNR felt lesser sleepy and found it easier to drive than the drivers in ConvR, but no difference of driver behaviour and driving performance was observed between two groups. In addition, regardless of the relationship, making casual conversation led decreased fatigue and workload in comparison with solo driving without conversation. The practical implication for further conversation methods to regard driver fatigue and driving performance in the automated vehicle is discussed.

Automated driving systems (ADS) in which human control is not required can only be achieved in limited and well predictable environments. Thus far, ADS users are expected to contribute to the automated process when requested by the system, during critical situations, and to ensure adequate system performance. Using a driving simulator experiment, this study developed a human-machine interface (HMI) in which the drivers can share decisions and control with the ADS to avoid slow traffic on highways. Following within-subjects design, the participants encountered four ADS designs in a randomized order: 1) the driver may regain the vehicle control and avoid the slow traffic manually without ADS support (no sharing); 2) an ADS requests the driver to resume the vehicle control and avoid the slow traffic manually (sharing decisions); 3) an ADS avoids the slow traffic autonomously if the driver approves it (sharing decision and control); and 4) an ADS avoids the slow traffic autonomously without driver's approval. Results indicated that the drivers preferred the first and second ADS designs more than the third and fourth ADS designs. However, significantly better system usability, drivers' engagement, and driving performance have been achieved by the third ADS design compared to other ADS designs. These results have implications for the safe implementation of partial and conditional driving automation systems.

Partially automated vehicles require human drivers to continuously monitor system status and external traffic environment. The `Out-of-The-Loop' phenomenon is a major issue with partially automated driving, and how to keep drivers remaining the driving loop and maintaining situation awareness for coping with system limits is an important question from perspective of human factors. In order to overcome fatigued driving caused by automation casual conversation with a passenger and a system was conducted, and to prevent the loss of situation awareness the information of traffic situation was presented to drivers. This study simulated two critical events: merging-car and curve road events to examine the impacts of conversation and the attention arousal. The statistical results found making conversation is significantly helpful in diminishing driver fatigue and boredom. Drivers' response to an impending collision on the merging-lane was quicker in driving during conversation with the system than with the passenger and solo driving, whereas no difference was observed in the curve event. The practical implication for further consideration with respect to the timing of attention arousal and the amount of information is discussed.

The paper aims to investigate how driving experience on system malfunction impacts driver take-over control on a long-term basis when he or she uses the conditionally driving automation. We designed a two-factorial (driving experience × knowledge of system limitation) driving simulator experiment at training and practicing phases between which there was a two-month period. All seventy-two people (36.2 ± 15.6 years old) participate into the data collection via driving simulators. The experimental results indicated that participants with the malfunction-experience at training reacted to a malfunction occurring at the two-month-later practicing phase. It is demonstrated that the experience was contributable for a long term. Also, it was shown that instructing the concept of the request to intervene (RtI) was attributed to rapidly taking-over car control. The effect of knowledge was significantly shown at the early training stage, and it was getting less as the driving experience was being accumulated. Furthermore, drivers' subjective assessment suggested that driver trust to driving automation would change with the accumulated experience. Meanwhile, the change was influenced by the acquired knowledge and experience at the training phase. Consequently, this study revealed the importance of experiencing system malfunction and instructing the knowledge at a learning/training period that will be helpful for drivers to achieve moderate driver-automation interaction.

Highly sophisticated and reliable driver assistance systems can fail if a driver does not fully understand the functionalities and capabilities of the system. A driving experiment was designed to evaluate how drivers' understanding of a driving assistance system and its authority impact on safety and performance in critical conditions. Two groups of 24 drivers each had to avoid several impending hazards, while each group received a different type of assistance for avoiding collisions during a lane change: a haptic feedback force through the steering wheel or an automatic steering control. The different assistance systems were applied to different hazards, resulting in significant differences in driver's reactions and behavior. Collision data showed that both systems were equally efficient when driver's expectations, system capabilities, and the hazard encountered were in line. More collisions were observed when both systems reached functional limits that were misunderstood by the drivers. However, the impact of these limitations varied, depending on the driver's ability to retake control and recover the critical situation. The overall driver's subjective post-hazard assessments were significantly affected by the smoothness of control authority transfer and the types of critical conditions. This indicates the necessity of an adaptive automation that can strike a balance between the processing abilities of the human and the system, and hazardous contexts encountered.

The purpose of this study is to clarify how to explain the system limitations in terms of possibility of hazard detection failure under partial driving automation, which is the level 2 of driving automation defined by Society of Automotive Engineers (SAE). Different from level 3 of driving automation which is supposed to issue Request to Intervene (RTI) to the driver whenever system limitations, an RTI may not be issued by the level 2 system because it is possible that the level 2 system fails to detect a hazard or understand the situation accurately which system cannot deal with. Under level 2, a driver therefore needs to understand the situation accurately and intervene in driving operation when necessary. This study focuses on whether explanation on the traffic scenes where the driving automation system may fail to detect a hazard has an impact on driver intervention or not. An experiment is conducted with a driving simulator (DS) in which the host vehicle is equipped with a driving automation system at level 2. A between-participants design is employed in our experiment. We have 30 participants, half of them are explained the possible traffic scenes and the other half are not. The participants are expected to find out the hazard which the system cannot detect a hazard by itself and to intervene in driving operation. The results show that drivers who were explained the possible scenes can deal well when the automation fails. According to the results, it may be claimed that it is necessary for car makers/dealers to explain to drivers on possible scenes where the driving automation system may fail to detect a hazard.

Better human machine cooperation is based on human trust in the machine. It has not been established, however, the system design which assures the machine is trusted by human operators in an appropriate manner. This is because the notion of trust is still vague since it has several aspects. The paper proposes to adopt a human-machine cooperation framework that the authors have developed. Based on the human-machine cooperation framework, each aspect of trust is identified. Examples are given to help readers understand each aspect.

The aim of the present study was to investigate elderly driver's reaction/response performance in dealing with a given alert owing to system limitations and to demonstrate what kind of knowledge may contribute to interventions into partial automation among elderly drivers. A two-factorial (knowledge of limitations × age) driving simulator experiment was conducted to collect data during the transition from automated control to manual driving (n=48 participants each in an elderly group [mean age, 70.4 years] and non-elderly group [mean age, 42.5 years]). The results revealed that with increased age, reaction times changed relatively little, but response behavior (i.e., performance after a reaction) declined considerably, especially for critical hazards. On the other hand, the results also showed that providing instruction for some of typical situations would improve driving performance for both elderly and non-elderly drivers. In addition, the comments made in relation to failed and delayed interventions implied that inadequate descriptions of system limitations may cause a driver to rely too heavily on system capabilities. Hence, the results of the present study demonstrate the importance of providing adequate knowledge about system limitations to elderly drivers. These findings are expected to be helpful in discussions regarding how to provide adequate knowledge/information to drivers in consideration of age, driving capability, and traffic conditions.

This paper investigated how different types of automation interventions affect tactical driving behavior during lane changes. Previous research indicates that drivers trust and accept the automation as long as they have the final authority over the system. However, the need for automation to be allowed to act autonomously in some time-critical situations has also been observed. This emerged the need for a system that is able to provide more than one type of assistance function depending on the situation. How drivers trust, accept, and interact with such multiple assistance functions is crucial for system effectiveness in terms of safety. A total of 40 participants were assigned to drive in an advanced driving simulator and encounter several critical lane change scenarios under one of four driving modes: unsupported driving; haptic steering control system; automatic steering control system; and integrated steering control system that comprises both haptic and automatic steering control assistance. Measures of trust in and acceptance of the systems were recorded with questionnaires. The results showed that the integrated steering control mode was efficient in terms of safety and able to attenuate the negative effects of both haptic and automatic systems.

It is important to learn about the factors that affect a driver's willingness to rely on an automated driving system in order to maintain appropriate trust in the system. The purpose of this study was to investigate the relationships between the system behavior and a driver's trust by considering overtaking a scooter and passing a bicycle automatically. Using a driving simulator, we first investigated the peak driving speed, the peak lateral distance between the object and the test vehicle, and the first steering input timing of human drivers while overtaking or passing. Next, the obtained data were applied to the parameters to design an automated driving system. Finally, we investigated the effects of differences in system behavior on drivers' trust by varying the system parameters. The results demonstrated that drivers' trust varied according to these three parameters. Moreover, the analysis indicated that an automated driving system should maintain longer peak lateral distances and engage in steering manoeuvres earlier than human drivers would in order to maintain a subjectively appropriate lateral distance and steering manoeuvre start time, respectively. Furthermore, the peak driving speeds for overtaking or passing should be kept almost equal to the manual speeds of human drivers to maintain subjectively appropriate driving speeds.

Emerging technologies in the field of automatization meanwhile enable partially and highly automated vehicles in the aviation and automotive domains, where the human operators are assisted or (partially/temporarily) replaced in their tasks by advanced automation systems. Nevertheless due to technological limitations and ethical reasons, full autonomous vehicles in both domains might not be realizable in the near future. Instead system designs, which enable shared and cooperative guidance and control of vehicles by human operators and automation systems could be more feasible solutions. This paper sketches a common framework of shared and cooperative control that describes the two concepts not as different but as coinciding concepts for the shared intentionality, control and cooperation between humans and machines. A brief overview off developed shared and cooperative control designs in the aviation and ground vehicle domain is given.

In order to realize really useful Level 2 driving automation systems that can be used in urban areas, it is a vital issue to develop a human-machine interface that assures drivers are able to recognize when to intervene into control. Visual information may not be appropriate for this purpose because the driver is required to monitor the traffic environment and the system status continuously. In this paper, we propose a driving automation system that performs a kind of "avoiding" maneuver against a hazard upon detection of the hazard. We expect that such avoiding maneuver makes a driver understand that the system has detected the hazard. Missing of such avoiding maneuver suggests the detection failure. We conduct an experiment with a driving simulator in order to examine whether such avoiding maneuver is really useful for drivers to understand the system status. The results suggest that it is actually useful if drivers hold the steering wheel. If, on the other hand, driver's hands are off the steering wheel, it tended to be difficult to recognize the system's failure.

Visual field defects, which are partial losses of the visual field, may have an impact on driver behavior. They affect the driver's information processing and are a possible cause of traffic accidents. Some patients who suffer from optic diseases related to visual field defects try to compensate for their disability by more actively moving their head in order to look around more carefully. Compensation is a driving strategy used to maintain safe driving regardless of the driver impairment. Several studies have shown that compensation is effective at enhancing the driving performance. However, few studies have examined both the compensation of drivers who have advanced visual field defects, and a quantitative analysis of driver performance while driving under such compensation. For these reasons, by using a driving simulator and eyeglasses that can imitate advanced visual field defects, this study investigates driver compensatory head movements to better understand the extent to which they are effective in avoiding pedestrian collisions. The results indicate that active head movements are efficient at reducing the number of pedestrian collisions compared with driving without such compensation. However, they are insufficient in terms of collision avoidance when compared to driving without a visual impairment. In other words, the results imply that there are assorted unavoidable circumstances that occur even when the driver attempts to compensate their visual impairment as much as possible.

Even highly sophisticated and reliable driver assistance systems might encounter critical situations that are outside system design capacity. This can lead to a conflict in viewpoints between the driver and automation. To evaluate the effects of such conflicts on human-machine interaction and cooperation, overall performance, and safety, an experiment was designed to determine driver responses to different imminent hazards during lane change while receiving automation assistance. Using a driving simulator, two types of lane change collision avoidance systems were provided: a haptic shared control that consists of auditory and haptic force warnings and an automatic shared control that includes auditory warnings and autonomous action. While the results of both systems were encouraging in terms of accident reduction when the hazards were within system design capacity, accidents were significantly increased when the hazards encountered were outside system design parameters. The acceptance by drivers was considerably influenced by the hazard type, and their feeling of control was affected by the type of assistance systems.

Based on current technologies, automated driving systems needs the human driver to monitor the environment and the machine. One way to enhance the driver to watch the environment is to make him/her put his/her hands on the steering wheel. Under the condition that the human collaborates with the automation, it is important for the automation to be designed that the automation is trustworthy. One of the issues to be addressed is a lane change for avoiding a forward obstacle. In this paper, we discuss the timing of starting lane change maneuver. Three types of the timing, i.e., Earliest, Mean, and Latest, were distinguished based on a preliminary data collection. The results showed that the three types would be acceptable, but Mean would be the best among the three types.

Driving with shared control systems has been shown to be efficient in reducing collisions during a lane change. However, the ability of the current shared control systems is limited and their absolute reliability is not assured. The goal of this study is to determine how changes in system performance affect human-machine cooperation and safety. An experiment using a driving simulator was designed to compare two types of support systems, (1) haptic shared control that provides control guidance to resists hazardous lane changes; and (2) mixed-input control that autonomously controls vehicle directions to prevent lane change collisions. Both systems were examined under three conditions: (i) 100% reliability, (ii) full reliability but limited system ability, and (iii) system failure. While both systems were judged trustworthy and accepted by the drivers when fully reliable, the mixed-input was more efficient at maintaining safety. In the second condition, drivers overestimating the capabilities of the system led to more collisions for both systems. The results illustrate that how machine failure impacts surprises and distrust in automation is highly dependent on driver's ability to regain the control. Overall, the haptic guidance was more efficient and accepted more by the drivers.

Driver drowsiness is a common cause of fatal traffic accidents. In this study, a driver assistance system with a dual control scheme is developed; it attempts to perform simultaneously the safety control of the vehicle and identification of the driver's state. The assistance system implements partial control in the event of lane departure and gives the driver the chance to voluntarily take the action needed. If the driver fails to implement the steering action needed within a limited time, the assistance system judges that “the driver's understanding of the given situation is incorrect” and executes the remaining control. We used a driving simulator equipped with the assistance system to investigate the effectiveness of identifying driver drowsiness and preventing lane departure accidents. Twenty students participated in three trials on a straight expressway, and they were required to implement only lateral control. We hypothesized that a participant cannot implement the action needed to maintain safety when he/she falls asleep, and that in such a case, the assistance system will implement the safety control repeatedly. The assistance system assisted the participants only when almost really needed, such as when their eyelids were closed. The results validated the hypothesis, showing that the assistance system implemented the safety control repeatedly when a participant fell asleep. In addition, the algorithms used by the assistance system to determine whether the driver can continue driving were evaluated through a leave-one-out cross-validation, and they were proven to be effective for identifying driver drowsiness.

This study investigates limitations of driver's compensation under visual field contraction. Two types are distinguished for the compensation: (A) reducing the vehicle speed, and (B) looking around frequently. An experiment using a driving simulator is conducted to evaluate to what extent accidents can be avoided by compensation. The participants are healthy drivers wearing eyeglasses that contract the participant's field of view. The results show that even if drivers perform compensation under visual field contraction as much as possible, they do not react early like when they drove normally without visual field contraction because they delay in perception of traffic condition. The accident rate is also high under the condition of visual field contraction. These findings suggest the need to design an assistance system that enhances drivers' ability to perceive traffic condition for people with visual field contraction.

The aim of this study was to investigate perspectives for securing driver's trust in automated driving. We considered driving scenes of passing other traffic objects (a bicycle or a scooter) and the relationship between the drivers' trust and the variety of speeding, taking lateral distance and steering timing by automated driving. A driving simulator study was implemented to investigate effects of applying characteristics of manual driving for each driver into design on drivers' trust when passing the objects. The results showed that driver's trust was affected by driver subjective speed, lateral distance and steering timing for automated driving. A longer lateral distance and an earlier steering timing for automated driving compared to manual driving contributed to subjectively appropriate speed and steering timing. Moreover the automated passing speed was regarded as an appropriate one if automated vehicles take the same or slightly slower speed compared to manual driving.

Goals of shared control usually focus on the sharing of the action. Several examples such as shared control in robotics, in car driving, in wheelchair control highlight the interest for Human operator to be able to take part in the control of a process in the same way as assistance systems. Moreover, with the haptic shared control, Human operator can feel what could be or what is the current action of the assistance system. But if shared control is so useful today in some domains and will also be very interesting for other domains, it is because action is not the only function that Human operator can share with an assistance system. In fact, we would even like to suppose that the more Human operator and assistance systems share problem solving process functions, the more their cooperation would be safe, efficient, and comfortable. The objective of this paper is to propose an extension of the use of Shared control concept by giving other opportunities for Human operator and assistance system to share activity. The horizontal extension relates the integration of all the functions, from information gathering to action implementation. The vertical extension relates the integration of the levels of activity usually represented by the strategic, tactical and operational levels. Human machine cooperation principles are used to attempt such an objective.

Graphical interface use involves schemata operations that range from transfer to induction. The former apply existing knowledge, such as prior schemata, and are effortless, preconscious and intuitive. The latter, which consist in constructing new schemata, are resource-consuming and thus detrimental to intuitive use (IU). A quantitative method is proposed to manipulate and screen schemata operations at the level of an interface's states and features. Relevance for the design cycle of innovative interfaces is critically reviewed, and integration with existing intuitive-use design frameworks is proposed. These considerations are built upon instructional design studies suggesting that assessment should precede and inform the application of design techniques geared toward IU.

This paper proposes a haptic assistance system that provides momentary steering wheel torque to help the driver choose the steering direction for avoiding collision with a pedestrian. The results of an experiment with a driving simulator showed that reaction time was reduced significantly when using haptic assistance and that the system was effective in enhancing appropriate selection of a steering direction in many cases. However, there were cases in which the drivers' choice was opposite to the system's proposal. In order for the system to be acceptable to drivers, the drivers' natural choice of direction should be taken into account.

This paper proposes a driver assistance system with a dual control theoretic scheme, which tries to perform safety control of a vehicle as well as identification of the driver's cognitive state simultaneously. The dual control theoretic driver assistance algorithm uses steering torque control based on a linear quadratic regulator (LQR). A controller and mathematical models for the vehicle and steering system were built, where the steering torque control was applied to switching feedback gains in the LQR controller. This paper discusses the dynamic characteristics for steering torque control under constraints on the state and manipulated variables.

This paper proposes a novel driver assistance system for enhancing driver recognition of vehicles in blind spots when executing lane changes. The system changes the yaw angle of the host vehicle's side-view mirror when another vehicle is in the blind spot. There are two design alternatives: (1) the system dynamically changes the side-view mirror angle or (2) the system turns on a lamp in addition to dynamically changing the side-view mirror angle. A cognitive experiment with a driving simulator was conducted for investigating system efficacy and influence on situation awareness of the driver. The results of this experiment show that the lamp may not be necessary and that both systems were effective when another vehicle was in the host vehicle's right blind spot.

This paper discusses an assistance system that add tiny torque to the steering wheel momentarily to help the driver choose the direction for avoiding collision with a pedestrian. In order to investigate the effectiveness of the system, we conducted an experiment with a medium fidelity driving simulator. The results of the experiment showed that there existed situations where the system was effective to enhance driver's appropriate selection of steering direction. In addition, the system reduced the driver reaction time significantly. However, there were cases in which the drivers completely disagreed with the system's proposal. In order to make such a system be acceptable to drivers, drivers' way of choosing a direction should be taken into account.

The authors investigated effectiveness of a pedestrian-vehicle collision avoidance system that was strictly the compatible with the human-centered automation. The system is designed to perform no avoidance maneuver when the driver does not show his/her intention to avoid a collision. This paper proposes a collision avoidance system that may intervene into control autonomously when the driver fails to show the intent in a highly emergency situation. The results of an experiment with a medium fidelity driving simulator show that the proposed system is effective to reduce the number of crashes that the original one can not prevent. Moreover, analyses on subjective ratings on improvement of collision avoidance capability and appropriateness of the system's maneuver show that the drivers have accepted the system.

Intuitive use is met when prior knowledge is transferred to new task environments. The empirical fact that transfer relies on schemas led us to diagnose intuitiveness based on schema induction. Two cognitive tasks were designed to make novice users perceive versus induce all the states of a prototype onboard computer. Subsequent interaction performances with the system validated the induction effect of the procedure and its interaction with familiarity, known as a primary factor of intuitive use. Implications for the diagnosis and the design of intuitive interfaces are discussed.

This article gives a microworld approach to identify design requirements for better situation awareness. Two experiments were designed and conducted. In the first experiment, behaviors of operators who lost situation awareness were analyzed, and the following two findings were obtained: (a) Automation must provide an operator with feedback information on the control mode, even when it is the operator who changed a control mode. (b) Authority for control may have to be passed from an operator to automation for attaining safety in highly urgent situations. The second experiment was done to investigate how human-interface may affect an operator's situation awareness. It is shown that human-interface must be carefully designed to externalize a mental model of the controlled process.

The third generation mobile radio system IMT-2000 CDMA TDD has been specified in the Third Generation Partnership Project (3GPP) with IMT-2000 CDMA TDD. This paper presents our TDD test-beds specifications receiving technology, downlink diversity, and open loop transmit power control (OL-TPC) as well as some field and our laboratory experiment results. According to our laboratory experiments, the joint detection can achieve better performance than RAKE receiver in case of two or more user transmission. The selective transmit diversity (STD) type transmit diversity (TX-Div) improves the BER performance around 4 dB compared to case without diversity. As the results of our field experiments, we acquired the characteristics of OL-TPC and a BER. We confirmed the feasibility of the CDMA TDD system from these results.

We examine the dependence of risk compensatory human adaptation to a system to attain safety of everyday machines, such as cars, on the level of automation and its shift. Two experiments were conducted using a microworld that is closely related to driving. The first experiment clarifies differences in the operation method and in risks involved in operation among different levels of automation. The results show that risk compensatory adaptation is significant if the computer of the safety system not only provides warning, but also performs safety control actions when necessary. In addition, even if the automation functions properly, new types of accidents arise, so that the number of accidents may not decline. The second experiment examines the dependence of human adaptation to a safety system on the direction of level shift of automation. The results show that the risk reducing adaptation when the level of automation decreases is not as significant as the risk compensatory adaptation when the level of automation increases.

This paper discusses a safety monitoring system design that consists of sensors with gray zone signal outputs, where the gray zone signal expresses no judgement for situations of uncertainty. This type of sensor is expected to prevent model-induced failures that may be induced by a failure to alarm or by generating false alarms, since the model-induced failures occur in the gray zone. Our problem of interest is in determining the optimal judgement method for the gray zone. Firstly, one-sensor system with three-value outputs is analyzed, and under certain conditions the two-step judgement method utilizing information from another sensor only for gray zone outputs are shown to be better than both deterministic and stochastic judgement. Secondly, a heterogeneous type of sensor system model is introduced where a judgement of safety or danger is always made after integrating information from a two-value sensor that monitors another characteristic of the real state. Lastly, it is suggested that these monitoring models are applicable to the analysis of the human collaboration problem.

In this paper, the possibility of signal transmission through a chain of Chua's circuits, each one operating in a chaotic-regime, is investigated. The propagation of an analog signal through a nearly synchronized chain is studied. As we increase the linear coupling resistance between cells, the chaotic noise is found to increase and propagate through the cells. This phenomenon is exploited in a new method for transmitting digital signals with a noise reduction circuit. We also present results on signal transmission delay. One of the most interesting results reported is our demonstration that it is possible under certain conditions to recover a binary signal even if no synchronization exists between cells.

A new method is given, which converts a chaotic motion in Chua's circuit to a periodic motion. A tunnel mechanism is used to perform this conversion.>

In this paper, we demonstrate how Yamakawa's chaotic chips and Chua's circuits can be used to implement a secure communication system based on chaotic modulation/demodulation systems. Furthermore, their performance for the secure communication is discussed.>