Udara E. Manawadu

Analysis of individual driving experience in autonomous and human-driven vehicles using a driving simulator

Intelligent vehicles capable of autonomous driving will be commercially available in near future. To formulate a beneficial relationship between driver and vehicle, it is important to analyze how the drivers would react to autonomous vehicles compared to human-driven (conventional) vehicles. In this study, we focused on analyzing individual driving experience in several road conditions for autonomous and conventional vehicles among experienced and novice drivers. We first developed a simplified driving simulator that can connect arbitrary interfaces, create virtual environments consisting of scenarios and events that drivers encounter in real-world driving, and implement fully autonomous driving. We then conducted experiments to clarify differences of driving experiences for autonomous driving between the two groups. The experimental results showed that experienced drivers opt for conventional driving mainly due to the flexibility and driving fun it offers, while novices tend to prefer autonomous driving due to its inherent easiness and safety. An in-depth analysis indicated that drivers preferred to use both the driving methods interchangeably depending on the road and traffic conditions.

A haptic feedback driver-vehicle interface for controlling lateral and longitudinal motions of autonomous vehicles

Autonomous vehicles will significantly change the existing driver-vehicle relationship, since only a destination input from the human driver will suffice. However, reduced degree of human-control could result in lack of driving pleasure and excitement. Thus, we proposed a method to increase the flexibility in controlling of an autonomous vehicle by allowing the driver to control its lateral and longitudinal motions with a time lag. We first derived a set of vehicle movements to improve driver experience and related them to a set of control functions that the driver can input. We then created two types of driver-vehicle interfaces (DVIs) for vehicle control; a haptic interface with kinesthetic and tactile feedback, and a hand-gesture interface with augmented reality feedback. The joystick-type haptic interface provides feedback on driver input by dynamically varying its degrees of freedom through controlling the current supplied to axis motors, and through vibration motors. The gesture interface is based on Leap Motion controller and provides visual feedback to driver. We conducted driving experiments in a VR simulator using twenty drivers to evaluate the effectiveness of these DVIs. The results showed that haptic interface significantly reduced the average input time and input error, and drivers preferred the haptic interface due to its ability to provide immediate, active, and passive feedback.

Estimating Driver Workload with Systematically Varying Traffic Complexity Using Machine Learning: Experimental Design

Traffic complexity is one of the factors affecting driver workload. In order to study the relationship between traffic complexity levels and workload, a designed experiment is required, especially to vary traffic flow parameters systematically in a simulated environment. This paper describes the experimental design of a simulator study for developing a computational model to estimate the behavior of driver workload based on traffic complexity. Driving simulators allow creating and testing different traffic scenarios and manipulating independent variables to improve the quality of data, as compared to real world experiments. Physiological responses such as heart rate, skin conductance, and pupil size have been found to be related to workload. By adapting a data-driven method, we integrated electrocardiography sensors, electro-dermal activity sensors, and eye-tracker to acquire driver physiological signals and gaze information. Preliminary results show a positive correlation between traffic complexity levels and corresponding physiological responses, performance, and subjective measures.

A multimodal human-machine interface enabling situation-adaptive control inputs for highly automated vehicles

Intelligent vehicles operating in different levels of automation require the driver to fully or partially conduct the dynamic driving task (DDT) and to conduct fallback performance of the DDT, during a trip. Such vehicles create the need for novel human-machine interfaces (HMIs) designed to conduct high-level vehicle control tasks. Multimodal interfaces (MMIs) have advantages such as improved recognition, faster interaction, and situation-adaptability, over unimodal interfaces. In this study, we developed and evaluated a MMI system with three input modalities; touchscreen, hand-gesture, and haptic to input tactical-level control commands (e.g. lane-changing, overtaking, and parking). We conducted driving experiments in a driving simulator to evaluate the effectiveness of the MMI system. The results show that multimodal HMI significantly reduced the driver workload, improved the efficiency of interaction, and minimized input errors compared with unimodal interfaces. Moreover, we discovered relationships between input types and modalities: location-based inputs-touchscreen interface, time-critical inputs-haptic interface. The results proved the functional advantages and effectiveness of multimodal interface system over its unimodal components for conducting tactical-level driving tasks.

Objective evaluation of oral presentation skills using Inertial Measurement Units

Oral presentation is considered as one of the most sought after skills by companies and professional organizations and program accreditation agencies. However, both learning process and evaluation of this skill are time demanding and complex tasks that need dedication and experience. Furthermore, the role of the instructor is fundamental during the presentation assessment. The instructor needs to consider several verbal and nonverbal communications cues sent in parallel and this kind of evaluation is often subjective. Even if there are oral presentation rubrics that try to standardize the evaluation, they are not an optimal solution because they do not provide the presenter a real-time feedback. In this paper, we describe a system for behavioral monitoring during presentations. We propose an ecological measurement system based on Inertial Measurement Units to evaluate objectively the presenters posture through objective parameters. The system can be used to provide a real-time feedback to the presenters unobtrusively.