Pengjun Zheng

A usability evaluation toolkit for In-Vehicle Information Systems (IVISs).

Usability must be defined specifically for the context of use of the particular system under investigation. This specific context of use should also be used to guide the definition of specific usability criteria and the selection of appropriate evaluation methods. There are four principles which can guide the selection of evaluation methods, relating to the information required in the evaluation, the stage at which to apply methods, the resources required and the people involved in the evaluation. This paper presents a framework for the evaluation of usability in the context of In-Vehicle Information Systems (IVISs). This framework guides designers through defining usability criteria for an evaluation, selecting appropriate evaluation methods and applying those methods. These stages form an iterative process of design-evaluation-redesign with the overall aim of improving the usability of IVISs and enhancing the driving experience, without compromising the safety of the driver.

To twist or poke? A method for identifying usability issues with the rotary controller and touch screen for control of in-vehicle information systems

In-vehicle information systems (IVIS) can be controlled by the user via direct or indirect input devices. In order to develop the next generation of usable IVIS, designers need to be able to evaluate and understand the usability issues associated with these two input types. The aim of this study was to investigate the effectiveness of a set of empirical usability evaluation methods for identifying important usability issues and distinguishing between the IVIS input devices. A number of usability issues were identified and their causal factors have been explored. These were related to the input type, the structure of the menu/tasks and hardware issues. In particular, the translation between inputs and on-screen actions and a lack of visual feedback for menu navigation resulted in lower levels of usability for the indirect device. This information will be useful in informing the design of new IVIS, with improved usability. Statement of Relevance: This paper examines the use of empirical methods for distinguishing between direct and indirect IVIS input devices and identifying usability issues. Results have shown that the characteristics of indirect input devices produce more serious usability issues, compared with direct devices and can have a negative effect on the driver–vehicle interaction.

In-Vehicle Information Systems to Meet the Needs of Drivers

In-Vehicle Information Systems (IVISs) integrate most of the secondary functions available within vehicles. These secondary functions are aimed at enhancing the driving experience. To successfully design and evaluate the performance of these systems, a thorough understanding of the task, user, and system is required. This article presents a review of these three variables in the context of IVISs, which aims to enhance understanding of this specific task–user–system interaction. A framework for modeling system performance for the task–user–system interaction is also proposed. This will allow designers and evaluators of IVISs to make predictions about system performance and to design systems that meet a set of criteria for usable IVISs.

Context of use as a factor in determining the usability of in-vehicle devices

In recent years, the issue of usability of in-vehicle devices has received growing attention. This is in line with the increase in functionality of these devices, which has been accompanied by the introduction of various new interfaces to facilitate the user–device interaction. The complexity and diversity of the driving task presents a unique challenge in defining usability: user interaction with in-vehicle devices creates a ‘dual task’ scenario, in which conflicts can arise between primary and secondary driving tasks. This, and the safety-critical nature of driving, must be accounted for in defining and evaluating the usability of in-vehicle devices. It is evident that defining usability depends on the context of use of the device in question. The aim of this review therefore is to define usability for in-vehicle devices by selecting a set of criteria to describe the various factors which contribute to usability in this specific context of use.

The effect of sensor errors on the performance of collision warning systems

Collision warning systems work on the information provided by different sensors. Among many factors that can affect the performance of collision warning systems, the effect of sensor errors is investigated. Errors in direct measurements and derivatives are analysed based on typical sensor configurations of collision warning systems, which include a laser speedometer, a Radar/Lidar. Empirical sensor outputs under real traffic situations are used in deriving distributions of measurement errors. The uncertainties of collision warning criteria are then examined based on empirically derived sensor errors for three collision warning algorithms, the Honda, Mazda and NHTSA algorithm. It is found that under some problematic traffic situations and for certain algorithms, the sensor errors could significantly affect the performance of collision warning systems.

Evaluation of Collision Warning-Collision Avoidance Systems Using Empirical Driving Data

Collision warning-collision avoidance (CWCA) systems are showing increasing promise in preventing rear-end collisions. The effectiveness of many prototype systems has been evaluated on driving simulators or through numerical simulation, and several field tests are under way. Most investigations have focused on the safety benefits to host vehicles. Possible effects of CWCA-equipped vehicles on other traffic, especially the direct follower, have received little attention. Results of an evaluation study of this issue are presented. It is found that CWCA vehicles may have higher risk of rear-end collision with following vehicles. Thus, it may be necessary to design a CWCA system that is able to offer a balanced protection to host vehicles and to other traffic by considering possible effects of CWCA systems on other traffic.

Effects of Intuitive Voice Interfaces on Driving and In-vehicle Task Performance

This paper reported results of an on-road evaluation study of three types of voice interfaces, the traditional voice system, and two intuitive voice systems with text prompts on a central display and on a cluster display respectively. The effects of voice interfaces were evaluated based on primary driving and secondary in-vehicle task performance. The in-vehicle performance was characterized by the mean task duration and mean error rates in performing in-vehicle operations. It was found that the mean task duration was the shortest in using the intuitive voice system with cluster display and drivers made more errors when using the traditional voice system relative to intuitive voice systems. The error rates were the lowest when using intuitive voice system with cluster display. The visual distraction effects were examined in terms of glance percentage to the windscreen (road ahead) and number of glances towards the central display and the cluster display. Reductions in glance percentage to the windscreen were observed when using intuitive voice interfaces, accompanied by increases in glance percentage to the prompt display. The primary driving performance in using the three voice-activation interfaces was not significantly affected compared with baseline car following only situation. It is concluded that intuitive voice interfaces are a viable enhancement to traditional voice interface whilst intuitive voice interface with cluster display has relative advantages of good task performance and minor visual distraction.

An algorithm for high-dimensional traffic data clustering

High-dimensional fuzzy clustering may converge to a local optimum that is significantly inferior to the global optimal partition. In this paper, a two-stage fuzzy clustering method is proposed. In the first stage, clustering is applied on the compact data that is obtained by dimensionality reduction from the full-dimensional data. The optimal partition identified from the compact data is then used as the initial partition in the second stage clustering based on full-dimensional data, thus effectively reduces the possibility of local optimum. It is found that the proposed two-stage clustering method can generally avoid local optimum without computation overhead. The proposed method has been applied to identify optimal day groups for traffic profiling using operational traffic data. The identified day groups are found to be intuitively reasonable and meaningful.

Evaluation of Effects of Ramp Metering on Merging Operations

Merging on motorways represents one of the most difficult and dynamic maneuvers of all driving subtasks. Merging operations at uncontrolled ramp junctions have been extensively investigated. With the rapid implementation of ramp metering techniques, many junctions are now operated under ramp metering control. Little research has been done to investigate the possible effects of ramp metering on the merging operation. The research described in this paper focuses on the comparisons of merging operations under both metered and unmetered scenarios. Time series data of the merging process were collected with the Transportation Research Group instrumented vehicle and video cameras at a tapered motorway merge site. The data included accurate vehicle speed measured by laser speedometer, leading and following headway measured by radar, and drivers eye movement derived from an in-car camera. Following a comprehensive analysis of the data, it was found that merging operations under ramp control could be significantly different from free merging. Merging maneuvers may become more difficult as a result of ramp control. The equity implications of such effects are discussed. It is believed that the understanding gained from this research will be useful for the design and operation of ramp metering.

Validation of capacity reductions in traffic monitoring systems

Both planned and unplanned events can reduce the capacities on roadways and result in congestion and delay. Currently, there are no complete and reliable sources which can be used by traffic operators to estimate capacity reductions. On many occasions, operators have to estimate capacity reductions using a large degree of subjectivity. This paper describes a methodology developed to validate capacity reductions from data collected within the traffic monitoring system. Methods for the determination of normal capacity, short-term and long-term capacity reductions are introduced. The proposed method has been successfully applied to derive capacity reductions using operational data. The results obtained are encouraging. It is believed that the method could be implemented in many traffic control centres to improve capacity reduction estimations