Amos Freedy

Measurement of trust in human-robot collaboration

We describe a collaborative performance model that captures the critical performance attributes of the distinctive human-robotic decision and control environment. The literature and our initial experimental studies show that the element of trust in human-robot collaboration is an extremely important factor in the performance model, and accordingly we have focused much of our attention on deriving suitable and practical measures of this variable. In this paper we describe the formulation of a decision-analytical based measure of trust as well as the results of two initial experiments designed to examine trust in a tactical human-robot collaborative task performed in our new mixed initiative team performance assessment system (MITPAS) simulation environment.

Electrotactile two-point discrimination as a function of frequency, body site, laterality, and stimulation codes.

The feasibility of frequency modulated two-point discrimination as a design concept for electrocutaneous sensory substitution display has been studied. Three stimulation techniques were tested on human subjects: spatial stimulus, temporal stimulus, and frequency on frequency stimulus. The frequency on frequency technique yielded the lowest threshold when compared to the temporal and spatial techniques. In addition, some of the characteristic behavior of cutaneous sensation is discussed relating two-point discrimination with frequency, body sites, and stimulation codes. Implications of the results for clinical applications are reviewed.

Knowledge Requirements and Management in Expert Decision Support Systems for (Military) Situation Assessment

Situation assessment tasks, e.g., medical diagnosis, battlefield reading, corporation status assessment for merger or acquisition purposes, are formulated as a general family of problem solving tasks. The generic nature of this family task as a multimembership hierarchical pattern recognition problem is characterized and the types of decision support systems (DSS) are identified. The focus is on knowledge representation and elicitation, although issues related to inference mechanisms, system structure, and expert-machine-user interface are also discussed. Two types of knowledge are distinguished: global knowledge and local knowledge. Global knowledge is required to determine directions on which to focus attention, while local knowledge is required for assessing the validity of a specific alternative based on a given set of findings. Global knowledge is represented as a network of relevancy pointers between alternatives and features. Attached to the links of this network are weights by which the strength of relevancy is evaluated and global directions (hypotheses) for situation analysis are determined. For local knowledge, it seems that in most practical problems multiple representation techniques would be required to characterize adequately the alternatives by means of their relevant features. The presentation is accompanied by examples for military situation assessment. However, comparable examples from medical and business applications are also cited. In fact, many of the ideas presented here have already been implemented in the MEDAS system¿a medical DSS for emergency and critical care medicine.

A Computer-Based Learning System for Remote Manipulator Control

A concept of adaptive aiding for performance improvement in remote handling is described. The concept incorporates an autonomous control subsystem (ACS) that is able to supplement the operators control function. The behavior of the ACS is established through a process of learning by observing the operators control function in relation to the environment and manipulator output. The computer-based system establishes a decision-making policy which is based on conditional probability. Initially, the output device is totally controlled by the operator, while the computer system acts as a passive observer. As the operation continues, the computer system gradually assumes the role of active controller, reducing the operators function to that of an action initiator and inhibitor. A pilot experiment indicates the feasibility of the concept; with a relatively short training period, the ACS was able to assume the bulk of the decision-making load and guide a three-dimensional manipulator satisfactorily through a series of manipulative tasks.

A Comprehensive Methodology for Assessing Human-Robot Team Performance for Use in Training and Simulation:

New methodologies and quantitative measurements for evaluating human-robot team performance must be developed to achieve effective coordination between teams of humans and unmanned vehicles. The Mixed Initiative Team Performance Assessment System (MITPAS) provides such a comprehensive measurement methodology. MITPAS consists of a methodology, tools and procedures to measure the performance of mixed manned and unmanned teams in both training and real world operational environments. This paper describes MITPAS and the results of an initial experiment conducted to validate the measures and gain insight into the effect of robot competence on operator trust as well as on human-robot team performance.

Designing an Adaptive Automation System for Human Supervision of Unmanned Vehicles: A Bridge from Theory to Practice:

Careful consideration must be given to the implementation of automation into complex systems. Much research in adaptive automation has identified challenges for system implementation. A key focus of this research has surrounded the methods of automation invocation including critical events, measurement, and modeling techniques. However, little consideration has been given to selecting and implementing appropriate techniques for a given system as a guide to designers of adaptive automation. This paper proposes such a methodology. We demonstrate the use of this methodology by describing a case study about a system designed to support effective communication and collaboration between the commander and vehicle operator in an unmanned aerial vehicle (UAV) system.

System integration of pattern recognition, adaptive aided, upper limb prostheses

Abstract An integrated externally energized artificial arm which combines computer control techniques and sensory feedback is presented. The arm, which is intended for above elbow amputees, contains the functions of prehension, wrist rotation, elbow flexion and extension and humeral rotation. The paper outlines the nature of control problems in artificial limbs and reviews techniques for their solutions. A detailed description of the system involving myoelectric pattern control, adaptive computer aiding and electrocutaneous feedback is included with a discussion of the integration and interfaces of these factors.

Adaptive Utility Assessment in Dynamic Decision Processes: An Experimental Evaluation of Decision Aiding

One central goal of decision theory is to provide a rational basis for decisionmaking. The ADDAM (Adaptive Dynamic Decision Aiding Methodology) system is designed to aid the decisionmaker (DM) in performing dynamic decision tasks. The ADDAM system provides real-time dynamic assessments of multiple utilities as the DM performs a dynamic decision task. ADDAM continuously tracks the DMs decision responses and uses adaptive pattern classification techniques to learn his utilities for their outcomes. These utilities are then used to provide decision aiding in the form of maximum expected utility decision recommendations. An experimental study was conducted to evaluate the effectiveness of the decision aiding system in a realistic decision task. Aided subjects showed significantly less deviation from their own maximum expected utility and substantially less within group-variance than did unaided subjects. Aided subjects also had a greater decision output. The adaptive utility estimates upon which the aiding was based converged rapidly to stable values. The ADDAM system was found to provide an appropriate, systematic, and testable approach to decision aiding. ADDAM aids the operator by organizing his own in-context decision behavior into a systematic mathematical framework. Such aiding is applicable to a wide variety of systems in which deficiencies of human decisionmakers may be overcome by techniques to augment human memory and logic processes.

Multiagent Adjustable Autonomy Framework (MAAF) for multi-robot, multi-human teams

This paper describes the ongoing development of a multiagent adjustable autonomy framework (MAAF) for multi-robot, multi-human teams performing tactical maneuvers. The challenge being addressed in this SBIR Phase I R&D project is how to exploit fully the unique capabilities of heterogeneous teams composed of a mixture of robots, agents or persons (RAPs): that is, how to improve the safety, efficiency, reliability and cost of achieving mission goals while maintaining dynamic adaptation to the unique limitations and contingencies of a real-world operating environment. Our response to this challenge is the creation of a new infrastructure that will facilitate cooperative and collaborative performance of human and robots as equal team partners through the application of advances in goal-oriented, multiagent planning and coordination technology. At the heart of our approach is the USC Teamcore Groups Machinetta, a state-of-the-art robot proxy framework with adjustable autonomy. Machinetta facilitates robot-human role allocation decisions and collaborative sharing of team tasks in the non-deterministic and unpredictable military environment through the use of a domain-independent teamwork model that supports flexible teamwork. This paper presents our innovative proxy architecture and its constituent algorithms, and also describes our initial demonstration of technical feasibility in a realistic simulation scenario.

Motivational factors in vigilance: effects of instructions on performance in a complex vigilance task.

Pre-task instructions have been a neglected source of motivation in vigilance. In the present study, 32 Ss monitored a complex visual vigilance display for 40 min. with a signal rate of 60/hr. 16 Ss were told that such tasks are usually challenging (positive set) and 16 were told such tasks are usually monotonous (negative set). Performance was significantly better throughout the session by Ss with the positive set and decrements did not occur with either group. The results indicate the importance of motivational factors in vigilance. Implications for vigilance research are discussed.