Nilanjan Sarkar

Control of Mechanical Systems With Rolling Constraints: Application to Dynamic Control of Mobile Robots

There are many examples of mechanical systems that require rolling contacts between two or more rigid bodies. Rolling contacts engender nonholonomic constraints in an otherwise holonomic system. In this article, we develop a unified ap proach to the control of mechanical systems subject to both holonomic and nonholonomic constraints. We first present a state space realization of a constrained system. We then dis cuss the input-output linearization and zero dynamics of the system. This approach is applied to the dynamic control of mo bile robots. Two types of control algorithms for mobile robots are investigated: trajectory tracking and path following. In each case, a smooth nonlinear feedback is obtained to achieve asymptotic input-output stability and Lagrange stability of the overall system. Simulation results are presented to demonstrate the effectiveness of the control algorithms and to compare the performance of trajectory-tracking and path-following algo rithms.

An empirical study of machine learning techniques for affect recognition in human-robot interaction

Given the importance of implicit communication in human interactions, it would be valuable to have this capability in robotic systems wherein a robot can detect the motivations and emotions of the person it is working with. Recognizing affective states from physiological cues is an effective way of implementing implicit human–robot interaction. Several machine learning techniques have been successfully employed in affect-recognition to predict the affective state of an individual given a set of physiological features. However, a systematic comparison of the strengths and weaknesses of these methods has not yet been done. In this paper, we present a comparative study of four machine learning methods—K-Nearest Neighbor, Regression Tree (RT), Bayesian Network and Support Vector Machine (SVM) as applied to the domain of affect recognition using physiological signals. The results showed that SVM gave the best classification accuracy even though all the methods performed competitively. RT gave the next best classification accuracy and was the most space and time efficient.

Adaptive control of an autonomous underwater vehicle: experimental results on ODIN

This paper presents a six-degrees-of-freedom controller for autonomous underwater vehicles. The control algorithm is adaptive in the dynamic parameters that are poorly known and time-varying in the underwater environment. Moreover, the proposed control law adopts quaternions to represent attitude errors, and thus avoids representation singularities that occur when using instead Euler angles description of the orientation. The adaptive controller has been successfully implemented and experimentally validated on omnidirectional intelligent navigator (ODIN), an autonomous underwater vehicle that has been designed and built at the University of Hawaii. The experimental results demonstrate the good performance of the proposed controller within the constraints of the sensory system.

Online Affect Detection and Robot Behavior Adaptation for Intervention of Children With Autism

Investigation into robot-assisted intervention for children with autism spectrum disorder (ASD) has gained momentum in recent years. Therapists involved in interventions must overcome the communication impairments generally exhibited by children with ASD by adeptly inferring the affective cues of the children to adjust the intervention accordingly. Similarly, a robot must also be able to understand the affective needs of these children-an ability that the current robot-assisted ASD intervention systems lack-to achieve effective interaction that addresses the role of affective states in human-robot interaction and intervention practice. In this paper, we present a physiology-based affect-inference mechanism for robot-assisted intervention where the robot can detect the affective states of a child with ASD as discerned by a therapist and adapt its behaviors accordingly. This paper is the first step toward developing ldquounderstandingrdquo robots for use in future ASD intervention. Experimental results with six children with ASD from a proof-of-concept experiment (i.e., a robot-based basketball game) are presented. The robot learned the individual liking level of each child with regard to the game configuration and selected appropriate behaviors to present the task at his/her preferred liking level. Results show that the robot automatically predicted individual liking level in real time with 81.1% accuracy. This is the first time, to our knowledge, that the affective states of children with ASD have been detected via a physiology-based affect recognition technique in real time. This is also the first time that the impact of affect-sensitive closed-loop interaction between a robot and a child with ASD has been demonstrated experimentally.

Online stress detection using psychophysiological signals for implicit human-robot cooperation

Robots are expected to be pervasive in the society in a not too distant future where they will work extensively as assistants of humans in various activities. With this in view, a novel affect-sensitive architecture for human-robot cooperation is presented in this paper where the robot is expected to recognize human psychological states. As a demonstration, an online heart rate variability analysis to infer the mental stress of a human engaged in a task is presented. This technique involves real-time heart rate monitoring, signal processing using both Fourier Transforrn and Wavelet Transform, and inferring the stress condition based on the level of activation of the sympathetic and parasympathetic nervous systems using fuzzy logic. Results from human subject trials are presented to validate the presented methodology. This stress detection technique is expected to be useful in the future human-robot cooperation activities, where the robot will recognize human stress and respond appropriately.

Dynamic Difficulty Adjustment in Computer Games Through Real-Time Anxiety-Based Affective Feedback

A number of studies in recent years have investigated the dynamic difficulty adjustment (DDA) mechanism in computer games to automatically tailor gaming experience to individual players characteristics. Although most of these existing works focus on game adaptation based on players performance, affective state experienced by the players could play a key role in gaming experience and may provide a useful indicator for a DDA mechanism. In this article, an affect-based DDA was designed and implemented for computer games. In this DDA mechanism, a players physiological signals were analyzed to infer his or her probable anxiety level, which was chosen as the target affective state, and the game difficulty level was automatically adjusted in real time as a function of the players affective state. Peripheral physiological signals were measured through wearable biofeedback sensors and several physiological indices were explored to determine their correlations with anxiety. An experimental study was conducted to evaluate the effects of the affect-based DDA on game play by comparing it with a performance-based DDA. This is the first time, that is known, that the impact of a real-time affect-based DDA has been demonstrated experimentally.

Dynamic control of 3-D rolling contacts in two-arm manipulation

When two or more arms are used to manipulate a large object, it is preferable not to have a rigid grasp in order to gain more dexterity in manipulation. It may therefore be necessary to control contact motion between the object and the effector(s) on one or more arms. This paper addresses the dynamic control of two arms cooperatively manipulating a large object with rolling contacts. In the framework presented here, the motion of the object as well as the loci of the contact point either on the surface of each effector or on the object can be directly controlled. The velocity and acceleration equations for three-dimensional rolling contacts are derived in order to obtain a dynamic model of the system. A nonlinear feedback control algorithm that decouples and linearizes the system is developed. This is used to demonstrate the control of rolling motion along each arm and the adaptation of grasps to varying loads.

Coordinated motion planning and control of autonomous underwater vehicle-manipulator systems subject to drag optimization

A new motion coordination algorithm for an autonomous underwater vehicle-manipulator system (UVMS) is proposed. This algorithm generates the desired trajectories for both the vehicle and the manipulator in such a way that the total hydrodynamic drag on the system is minimized. Resolution of kinematic redundancy of the system is performed at the acceleration level so that this algorithm can be incorporated into the system dynamics. The dynamics of the UVMS are modeled using a quasi-Lagrange approach. A state-space formulation of the system along with a model-based controller design for trajectory-following tasks that includes thruster dynamics is also presented. The computer simulation results demonstrate the effectiveness of this proposed method in reducing the drag on the system.

Physiology-based affect recognition for computer-assisted intervention of children with Autism Spectrum Disorder

Generally, an experienced therapist continuously monitors the affective cues of the children with Autism Spectrum Disorders (ASD) and adjusts the course of the intervention accordingly. In this work, we address the problem of how to make the computer-based ASD intervention tools affect-sensitive by designing therapist-like affective models of the children with ASD based on their physiological responses. Two computer-based cognitive tasks are designed to elicit the affective states of liking, anxiety, and engagement that are considered important in autism intervention. A large set of physiological indices are investigated that may correlate with the above affective states of children with ASD. In order to have reliable reference points to link the physiological data to the affective states, the subjective reports of the affective states from a therapist, a parent, and the child himself/herself were collected and analyzed. A support vector machines (SVM)-based affective model yields reliable prediction with approximately 82.9% success when using the therapists reports. This is the first time, to our knowledge, that the affective states of children with ASD have been experimentally detected via physiology-based affect recognition technique.

Unified formulation of robotic systems with holonomic and nonholonomic constraints

Many robotic systems are subject to nonholonomic as well as holonomic constraints. Rolling contact between two rigid bodies is a typical example of such a system. In the study, a unified state space formulation of robotic systems subject to both holonomic and nonholonomic constraints is presented. The position-level holonomic constraints are first replaced by a set of velocity-level constraint equations that asymptotically converge to the original holonomic constraints. Having represented both holonomic and nonholonomic constraints in a common form, a state space representation of the constrained systems is then developed. A numerical algorithm for implementing the state space representation is also described. The proposed formulation eliminates the need to solve holonomic constraints either analytically or numerically, and ensures that holonomic constraints are always satisfied, particularly in computer simulations. The formulation makes it possible to treat systems with holonomic constraints, with nonholonomic constraints, or with both holonomic and nonholonomic constraints in a unified framework. Two examples are presented to illustrate the application of the unified formulation.