PRASHANT JAMWAL - Profile on Academia.edu (original) (raw)

Uploads

Papers by PRASHANT JAMWAL

Expert Systems With Applications, 2011

Pneumatic muscle actuators (PMA) show great potential in wearable and compliant rehabilitation de... more Pneumatic muscle actuators (PMA) show great potential in wearable and compliant rehabilitation devices as they are flexible and lightweight. However, the varying and non-linear behavior of the actuators imposes modeling and control challenges, which are difficult to comprehend. This research proposes a new wearable ankle rehabilitation robot, first of its kind in the world driven by PMAs in a parallel form. The focus of this presented work is to develop an iterative controller to overcome the challenges for PMA driven devices. A fuzzy feedforward controller is proposed to accurately predict the behavior of PMA. A modified Genetic Algorithm (GA) is developed to identify the optimal set of parameters for the fuzzy controller. The iterative controller has been tested on the proposed PMA driven ankle rehabilitation robot, and is found capable of mapping the complex relationship in length, force and pressure of the PMA with high accuracy. Experimental results show excellent trajectory tracking performance of the controller when given various desired trajectories.

This paper describes the kinematics and the dynamics of a 3-DOF pneumatic muscle driven wearable ... more This paper describes the kinematics and the dynamics of a 3-DOF pneumatic muscle driven wearable parallel robot designed for ankle rehabilitation treatments. Several important performance indices are identified to accomplish the requirements of the ankle rehabilitation treatment and the wearable robot design. It is found that some of these indices are conflicting and hence in order to obtain an optimal robot design, these indices are required to be simultaneously optimized. Consequently, a multi-objective optimization scheme based on evolutionary algorithms is proposed in this paper to find a design which has optimum performance. The proposed design analysis can be generalized with modest efforts for the development of all the classes of parallel robots.

Robot human interaction requires use of safe, compliant and light weight actuators. Conventional ... more Robot human interaction requires use of safe, compliant and light weight actuators. Conventional linear motors and pneumatic cylinders are normally used to actuate robots to assist and augment human motions. Lately it has been realized that these actuators are not suitable and safe for applications involving human actor. Their large weight, size and stiffer design raise concerns. Pneumatic muscle actuators (PMA) on the other hand are very light weight, compact and have inherent compliance which make them potential candidate for applications involving robot human interaction. Taking on the advantages, these actuators are now being experimented for a variety of medical and rehabilitation applications. However they are not very popular due to their highly nonlinear and time dependent behavior which poses control problems. In this paper, an attempt is being made to accurately predict the uncertain and ambiguous characteristics of PMA using Artificial Intelligence (AI). Conventional tools such as analytical and numerical methods can only model a nonlinear system which is time independent. Time varying nonlinear system characteristics can be best modeled using artificial intelligence-based regression models. In this research, Artificial Neural Network (ANN), Mamdani Fuzzy Inference System (FIS) and Takagi-Sugeno (TS)-based fuzzy system are developed after carefully analyzing the time series data obtained from a real system. To achieve higher accuracy from these models, their parameters are tuned. Parameters of ANN are tuned using back propagation algorithm whereas fuzzy parameters are tuned using three different methods, namely, gradient descent method (GD), genetic algorithms (GA) and Modified Genetic Algorithm (MGA). It was found that the TS fuzzy inference system tuned by MGA provides better accuracy and can also model the time dependent behavior of PMA. The proposed TS fuzzy system is found to perform better in terms of accuracy and maximum deviation when compared to the previous approaches in the literature.

Dynamic modeling of pneumatic muscles using modified fuzzy inference mechanism

Page 1. Abstract—Pneumatic muscle actuators (PMA), owing to their obvious advantages over convent... more Page 1. Abstract—Pneumatic muscle actuators (PMA), owing to their obvious advantages over conventional linear actuators and pneumatic cylinders, have been recently used in the medical and industrial robotic applications. ...

Mechanism and Machine Theory, 2010

This article deals with forward kinematics (FK) mapping of a parallel robot, especially designed ... more This article deals with forward kinematics (FK) mapping of a parallel robot, especially designed for ankle joint rehabilitation treatments. Parallel robots exhibit highly coupled non-linear motions hence conventionally a unique closed form solution of their FK cannot be obtained. However, since FK is a key module in closed loop position and force control, its accurate and fast solution is indispensable. To solve the FK problem, a modified fuzzy inference system (FIS) is proposed in this paper for the first time which is time efficient and becomes very accurate when its parameters are optimized. In the proposed work, FIS has been optimized using three approaches namely: gradient descent (GD), genetic algorithm (GA) and modified genetic algorithm (MGA). The FIS, optimized by MGA has been found to be more accurate than the GD and GA optimized FIS. Performance of the MGA based fuzzy system has been found better both in terms of accuracy and computation time, when compared with Newton-Raphson iterative method and other fuzzy and neural approaches. j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / m e c h m t 1538 P.K. Jamwal et al. / Mechanism and Machine Theory 45 (2010) 1537-1554

Robotics and Autonomous Systems, 2009

Rehabilitation robotics is an evolving area of active research and recently novel mechanisms have... more Rehabilitation robotics is an evolving area of active research and recently novel mechanisms have been proposed to reinstate complex human movements. Parallel robots are of particular interest to researchers since they are rigid and can provide enough load capacity for human joint movements. This paper proposes a soft parallel robot (SPR) for ankle joint rehabilitation. Kinematic workspace analysis is carried out and the singularity criterion of the SPR's Jacobian matrix is used to define the feasible workspace. A global conditioning number (GCN) is defined using the Jacobian matrix as a performance index for the evaluation of the robot design. An optimization problem is formulated to minimize the GCN using modified genetic algorithm (GA). Results from simple GA and modified GA are compared and discussed. As a result of the optimization, an optimal robot design is obtained which has a near unity GCN with almost uniform distribution in the entire feasible workspace of the robot.

Expert Systems With Applications, 2011

Pneumatic muscle actuators (PMA) show great potential in wearable and compliant rehabilitation de... more Pneumatic muscle actuators (PMA) show great potential in wearable and compliant rehabilitation devices as they are flexible and lightweight. However, the varying and non-linear behavior of the actuators imposes modeling and control challenges, which are difficult to comprehend. This research proposes a new wearable ankle rehabilitation robot, first of its kind in the world driven by PMAs in a parallel form. The focus of this presented work is to develop an iterative controller to overcome the challenges for PMA driven devices. A fuzzy feedforward controller is proposed to accurately predict the behavior of PMA. A modified Genetic Algorithm (GA) is developed to identify the optimal set of parameters for the fuzzy controller. The iterative controller has been tested on the proposed PMA driven ankle rehabilitation robot, and is found capable of mapping the complex relationship in length, force and pressure of the PMA with high accuracy. Experimental results show excellent trajectory tracking performance of the controller when given various desired trajectories.

This paper describes the kinematics and the dynamics of a 3-DOF pneumatic muscle driven wearable ... more This paper describes the kinematics and the dynamics of a 3-DOF pneumatic muscle driven wearable parallel robot designed for ankle rehabilitation treatments. Several important performance indices are identified to accomplish the requirements of the ankle rehabilitation treatment and the wearable robot design. It is found that some of these indices are conflicting and hence in order to obtain an optimal robot design, these indices are required to be simultaneously optimized. Consequently, a multi-objective optimization scheme based on evolutionary algorithms is proposed in this paper to find a design which has optimum performance. The proposed design analysis can be generalized with modest efforts for the development of all the classes of parallel robots.

Robot human interaction requires use of safe, compliant and light weight actuators. Conventional ... more Robot human interaction requires use of safe, compliant and light weight actuators. Conventional linear motors and pneumatic cylinders are normally used to actuate robots to assist and augment human motions. Lately it has been realized that these actuators are not suitable and safe for applications involving human actor. Their large weight, size and stiffer design raise concerns. Pneumatic muscle actuators (PMA) on the other hand are very light weight, compact and have inherent compliance which make them potential candidate for applications involving robot human interaction. Taking on the advantages, these actuators are now being experimented for a variety of medical and rehabilitation applications. However they are not very popular due to their highly nonlinear and time dependent behavior which poses control problems. In this paper, an attempt is being made to accurately predict the uncertain and ambiguous characteristics of PMA using Artificial Intelligence (AI). Conventional tools such as analytical and numerical methods can only model a nonlinear system which is time independent. Time varying nonlinear system characteristics can be best modeled using artificial intelligence-based regression models. In this research, Artificial Neural Network (ANN), Mamdani Fuzzy Inference System (FIS) and Takagi-Sugeno (TS)-based fuzzy system are developed after carefully analyzing the time series data obtained from a real system. To achieve higher accuracy from these models, their parameters are tuned. Parameters of ANN are tuned using back propagation algorithm whereas fuzzy parameters are tuned using three different methods, namely, gradient descent method (GD), genetic algorithms (GA) and Modified Genetic Algorithm (MGA). It was found that the TS fuzzy inference system tuned by MGA provides better accuracy and can also model the time dependent behavior of PMA. The proposed TS fuzzy system is found to perform better in terms of accuracy and maximum deviation when compared to the previous approaches in the literature.

Dynamic modeling of pneumatic muscles using modified fuzzy inference mechanism

Page 1. Abstract—Pneumatic muscle actuators (PMA), owing to their obvious advantages over convent... more Page 1. Abstract—Pneumatic muscle actuators (PMA), owing to their obvious advantages over conventional linear actuators and pneumatic cylinders, have been recently used in the medical and industrial robotic applications. ...

Mechanism and Machine Theory, 2010

This article deals with forward kinematics (FK) mapping of a parallel robot, especially designed ... more This article deals with forward kinematics (FK) mapping of a parallel robot, especially designed for ankle joint rehabilitation treatments. Parallel robots exhibit highly coupled non-linear motions hence conventionally a unique closed form solution of their FK cannot be obtained. However, since FK is a key module in closed loop position and force control, its accurate and fast solution is indispensable. To solve the FK problem, a modified fuzzy inference system (FIS) is proposed in this paper for the first time which is time efficient and becomes very accurate when its parameters are optimized. In the proposed work, FIS has been optimized using three approaches namely: gradient descent (GD), genetic algorithm (GA) and modified genetic algorithm (MGA). The FIS, optimized by MGA has been found to be more accurate than the GD and GA optimized FIS. Performance of the MGA based fuzzy system has been found better both in terms of accuracy and computation time, when compared with Newton-Raphson iterative method and other fuzzy and neural approaches. j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / m e c h m t 1538 P.K. Jamwal et al. / Mechanism and Machine Theory 45 (2010) 1537-1554

Robotics and Autonomous Systems, 2009

Rehabilitation robotics is an evolving area of active research and recently novel mechanisms have... more Rehabilitation robotics is an evolving area of active research and recently novel mechanisms have been proposed to reinstate complex human movements. Parallel robots are of particular interest to researchers since they are rigid and can provide enough load capacity for human joint movements. This paper proposes a soft parallel robot (SPR) for ankle joint rehabilitation. Kinematic workspace analysis is carried out and the singularity criterion of the SPR's Jacobian matrix is used to define the feasible workspace. A global conditioning number (GCN) is defined using the Jacobian matrix as a performance index for the evaluation of the robot design. An optimization problem is formulated to minimize the GCN using modified genetic algorithm (GA). Results from simple GA and modified GA are compared and discussed. As a result of the optimization, an optimal robot design is obtained which has a near unity GCN with almost uniform distribution in the entire feasible workspace of the robot.