Priyanshu Agarwal

I am a Postdoc 
working with Dr. Marcia O'Malley 
in the Mechanical Engineering Department at Rice University. I completed my PhD 
working with Dr. Ashish Deshpande in the Mechanical Engineering Department 
I worked on the NSF and NASA funded hand exoskeleton project with applications in stroke rehabilitation and teleoperation.
In my Master's, I worked on the 
DARPA funded Mind's Eye Program, where I 
developed dynamics-based biomechanical and statistical motion models for tracking human pose in monocular videos as a part of the ISTARE research effort.

Research Interests:
Rehabilitation devices
Human Biomechanics
Dynamics and Control

M.S., Mechanical Engineering (Robotics & Computer Vision), The State University of New York at Buffalo (SUNY Buffalo), 2012 (GPA: 4.0/4.0)
B.Tech., Mechanical Engineering, Motilal Nehru National Institute of Technology, Allahabad (MNNIT Allahabad), India, 2007 (Gold-medalist)
XII Delhi Public School, Vidyut Nagar, 2003 (Rank: 1/100)

 Travel Award, Proposal: A Synergistic Approach to Development of Musculoskeletal Models and Rehabilitation Exoskeletons using Experimental Data, National Center for Simulation in Rehabilitation Research, Stanford University, 2016
 Best Paper Award: Technical Category, IEEE RO-MAN, 2016
 Graduate School Continuing Fellowship, UT Austin, 2016
 Best Robotics Paper Award, Dynamic Systems and Control Conference (DSCC), 2015
 GAIN Poster Award, Poster: UT Hand Exoskeleton for Rehabilitation and TeleoperationGraduate and Industry Networking, UT Austin, 2014
 Best Poster Award (First Place), Poster: A Novel Hand Exoskeleton for Rehabilitation and Teleoperation, Research Poster Contest, Department of Mechanical Engineering, UT Austin, 2014
 Best Poster Award (Second Place), Poster: Design and Development of a Hand-Wrist Exoskeleton for Stroke Rehabilitation, Research Poster Contest, Department of Mechanical Engineering, UT Austin, 2013
 Outstanding Researcher Award, Proposal: Virtual Design, Control, and Testing of Hand Exoskeletons using Simulation, National Center for Simulation in Rehabilitation Research, Stanford University, 2012
 National ICT Australia (NICTA) Student Fellowship, Robotics: Science and Systems Conference, 2012
 George J. Heuer, Jr. Ph.D. Endowed Graduate FellowshipCockrell School of EngineeringUT Austin, 2012-13
 Best Poster Award (First Place), Poster: Articulated Body Pose Estimation in Monocular Video, 23rd CSE Graduate Research Conference, SUNY Buffalo, 2011
 Most Promising Graduate Trainee Engineer (2007-08); Award for excellent performance (2007-08), R&D Bajaj Auto Ltd.
 Mohit Chaturvedi Memorial Gold Medal (standing first, 2007); Saroj Agarwal Gold Medal (standing first, 2006); Kiran Agarwal Gold Medal (scoring highest marks in Machine Design), B.Tech, Mechanical EngineeringMNNIT Allahabad
 Summer Fellow, Indian Institute of Technology, Madras, 2006 (20 candidates selected in India)

Media Links:
Alcalade Magazie (01/09/2014) 

Engineer (Design)                                      July' 2008 - June' 2009
Graduate Trainee Engineer,                         July' 2007- July' 2008
Transmission Division,                                      
**Ahead, Research & Development Department,
Bajaj Auto Ltd., Pune, India

State University of New York, Buffalo, NY

Teaching Assistant in Mechanical and Aerospace Engineering Department             August 09 - May 10

Courses Taught

MAE277: Introduction to Mechanical Engineering Practice                                     August 09 - December 09
MAE 340: Systems Analysis (Lab)                                                                            January 10 - May 10

Program Committee:
Robotics Science and Systems (C)

Review Editor: 
Frontiers in Robotics and AI (Robotic Control Systems)

Computer Methods and Programs in Biomedicine (J)
Control Engineering Practice (J)
ASME Dynamic Systems and Control Conference (C)
Frontiers in Neurorobotics (J)
Frontiers in Robotics and AI (J)
IEEE Access (J)
IEEE/ASME Transactions on Mechatronics (J)
IEEE International Conference on Automation Science and Engineering (C)
IEEE RAS/EMBS International Conference on Biomedical Robotics and Biomechatronics (C)
IEEE International Conference on Robotics and Automation (C)
IEEE International Conference on Soft Robotics (C)
IEEE Robotics and Automation Letters (J)
IEEE/RAS-EMBS International Conference on Biomedical Robotics and Biomechatronics (C)
IEEE International Symposium on Robot and Human Interactive Communication (C)
IEEE/RSJ International Conference on Intelligent Robots and Systems (C)
IEEE Transactions on Haptics (J)
IEEE Transactions on Robotics (J)
Indian Control Conference (C)
Information Sciences (J)
International Journal of Intelligent Robotics and Applications (J)
International Journal of Robotics Research (J)
Journal of NeuroEngineering and Rehabilitation (J)
Mechatronics (J)
Mechanical Systems and Signal Processing (J)
Robotics Science and Systems (C)
Sensors (J)

Research Highlights:
(Please refer to the Publications page for an updated list of publications)

A Hand Exoskeleton with Series Elastic Actuation for Rehabilitation

Rehabilitation of the hands is critical for restoring independence in activities of daily living for individuals with upper extremity disabilities. There is initial evidence that robotic devices with force-control based strategies can help in effective rehabilitation of human limbs. However, to the best of our knowledge, none of the existing hand exoskeletons allow for accurate force or torque control. We design and prototype a novel hand exoskeleton with the following unique features: (i) an underlying kinematic mechanism that is optimized to achieve large range of motion, (ii) Bowden-cable-based LC-SEA allowing for bidirectional torque control of each joint individually. We test the developed prototype with human subjects to characterize its kinematics, dynamics and controller performance. Results show that the device supports a large workspace with hand, preserves the characteristics of natural motion of each digit, allows for accurate torque control and can be rendered dynamically transparent to offer minimal resistance to digit motion.

Assist-as-needed Controllers for a Hand Exoskeleton for Rehabilitation
We present two types of subject-specific assist-as-needed controllers. Learned force-field control is a novel control technique in which a neural-network-based model of the required torques given the joint angles for a specific subject is learned and then used to build a force-field to assist the joint motion of the subject to follow a trajectory designed in the joint-angle space. Adaptive assist-as-needed control,on the other hand, estimates the coupled digit-exoskeleton system torque requirement of a subject using radial basis function (RBF) and on-the-fly adapts the RBF magnitudes to provide a feed-forward assistance for improved trajectory tracking. Experiments on the index finger exoskeleton prototype with a healthy subject showed that while the force-field control is non-adaptive and there is less control on the speed of execution of the task, it is safer as it doesnot apply increased torques if the finger motion is restricted. On the other hand, adaptive assist-as-needed controller adapts to the changing needs of the coupled finger-exoskeleton system and helps in performing the task with a consistent speed, however, applies increased torques in case of restricted motion and therefore, less safe.

A Novel Framework for Optimizing Motor (Re)-learning using a Robotic Exoskeleton 

A critical question to be answered to improve robotic rehabilitation is what is the optimal rehabilitation environment for a subject that will facilitate maximum recovery during therapy? Studies suggest that task variability, nature and degree of assistance or error-augmentation and type of feedback play a critical role in motor (re)-learning. In this work, we present a framework for robot-assisted motor (re)-learning that provides subject-specific training by allowing for simultaneous adaptation of task, assistance and feedback based on the performance of the subject on the task. We model a continuous and coordinated multi-joint task using a learning from demonstration approach, which allows the task to be modeled in a generative manner such that the challenge-level of the task could be modulated in an online manner. To train the subjects for dexterous manipulation, we present a torque- based task that requires subject to dynamically regulate their joint torques. Finally, we carry out a pilot study with healthy human subjects using our previously developed hand exoskeleton to test a hypothesis and the results suggest that training under simultaneous adaptation of task, assistance and feedback positively affects motor learning.

Musculoskeletal Analysis as a Tool to Study Design/ Control of Upper-Limb Exoskeletons 

Exoskeletons are a new class of articulated mechanical systems whose performance is realized ... Read More

  1. P. Agarwal, M. S. Narayanan, L-F. Lee, F. Mendel and V. N. Krovi, "Simulation-based Design of Exoskeletons Using Musculoskeletal Analysis", proceeding of ASME 2010 International Design Engineering Technical Conference, 2010. (  CIE Best Paper Award) [PDF][PPT][IDETC Talk]

An Optimization Based Framework for Human Pose Estimation in Monocular Videos  ()

Human pose estimation using monocular vision is a challenging problem in computer vis ...Read More

  1. P. Agarwal, S. Kumar, J. Ryde, J. Corso, and V. Krovi, "An Optimization Based Framework for Human Pose Estimation in Monocular Videos", International Symposium on Visual Computing, Rethymnon, Crete, Greece, July 16-18, 2012. [PDF] ( Best Poster Award, 23rd CSE Graduate Research Conference, State University of New York at Buffalo, 2011.)
  2. P. Agarwal, "An Optimization Framework for Pose Estimation of Human Lower Limbs from a Singe Image", Technical report, SUNY, Buffalo. [Report] [Poster

Dynamics-based Biomechanical and Statistical Human Motion Modeling for People Tracking (DARPA Mind's Eye Program)

Human tracking is one of the most challenging problems in computer ...Read More

1. P. Agarwal, S. Kumar, J. Ryde, J. Corso, and V. Krovi, "Estimating Human Dynamics On-the-fly Using Monocular Video for Pose Estimation", Robotics: Science and Systems Conference, University of Sydney, Sydney, Australia, July 9-13, 2012. [PDF]

2. P. Agarwal, S. Kumar, J. Corso, and V. Krovi, "Estimating Dynamics On-the-fly Using Monocular Video", Dynamic Systems and Control Conference, California, October 12-14, 2011. [PDF][PPT]

Youtube Channel: 

SMARTY: A Vision Based Autonomous Ball Sorting Robotic System

Autonomous mobile robots capitalize on their ability to intelligently, efficiently and reliably interact with their environment. This makes them suitable for various applications that are hazardous or unpleasant for human beings. Their task may range from as simple as following a line, to as complex as ....Read More      


Presentation              Project Report                     Source Code

Backstepping Torque Control for Cylindrical Cam in Sequential Type Gearshift Mechanism

Faster and efficient gear shifts have always been the demand of automobile industry. Be it an automatic transmission having distinctly low efficiency or an automated manual transmission with combined benefits of automatic and manual gearboxes, the phenomenon of gear shift has always been the area of interest. Sequential type gear shift mechanisms  ... Read More

             Presentation1     Presentation 2     Presentation 3


                                        Project Report

The objective of this project is to develop a model for the dynamics of Sequential type gearshift mechanism and to study the controllability and observability of the mechanism as a dynamic system. A mathematical model capable of ... Read More

Project Report

Kinematic Modeling and Analysis of 3-PRR Parallel Manipulator using MATLAB GUI 

The analysis of 3-PRR manipulators is carried out. Forward and Inverse pose kinematics of the manipulator is derived. The Jacobian matrix relating the task space velocity to the join-space velocity is then symbolically evaluated. Various algorithms are... Read More

 Project Report                     MATLAB Source Code

( Showcased in the MATLAB kiosk as one of the available tools to better understand modeling of parallel manipulators in MATLAB, IEEE International Conference on Robotics and Automation, 2012.)

The problem of gearshift control is crucial for effecting an efficient gearshift. The issue becomes more interesting when it is constrained by yet another demand i.e. reduced gearshift time. In this paper, we address the issue of designing a controller for a sequential type gearshift mechanism in order to achieve the dynamically varying demand of gearshift force. A difference equation governing the dynamics of the gearshift mechanism is derived ...Read More

P. Agarwal, "Discrete-time Backstepping Torque Controller for Cylindrical Cam in Sequential type Gearshift Mechanism", Project Paper, Optimization in Engineering Design. [Paper]

An Electronic Differential based Electric Vehicle

P. Agarwal, G. Sharma, D. Ghosh, M. Satish, A. Upadhay, "Design and Development of a Battery Operated Electric Car with an Electronic Differential", B.Tech Project Report, MNNIT Allahabad, 2007. [Report]

Simulation Aided Teaching: Character Recognition, Graph Generation and Robot Control

P. AgarwalR. Gupta, P. K. Chaurasia, “Development of a system for Character Recognition, Graph Generation, and Autonomous Robot Control,” All India Seminar on Advances in Product Development, MNNIT, Allahabad, Feb 2006. [PDF]

A system for Dynamically Balancing a Bicycle 

(Summer Fellowship Programme of  Indian Institute of Technology, Madras, India under the guidance of Professor Nilesh J. Vasa.)

A study of the dynamics of a bicycle is carried out to evaluate the effect of key parameters like wheel base, head angle and trail on stability. It is observed that the design of front fork has a major impact on the stability of the bicycle. Second-order models are developed for the bicycle and a transfer function is finally established relating the roll angle to the steer angle. A PID controller is then modeled in MATLAB using Simulink. The response of the controller for various inputs like impulse, step etc. is then obtained. A frequency response study is then carried out for sinusoidal inputs using Bode plot and Nyquist Criterion.

         Automated Guided Vehicle for Mixed Model Assembly Line

A computer controlled automated guided vehicle is developed with a capability (1) to identify balls of different colors, (2) to place them at their appropriate location based on their color. The vehicle is a white line follower installed with two sensors for guiding it to track the line. In addition to this, the vehicle is equipped with a sensor cluster capable of identifying few colors. The sensor is based on the theory that different colors have different reflectivity. All the sensors are based on a circuit consisting of a Light Dependent Resistor connected in series with a rotary type potentiometer so as to generate a voltage output based on the reflectivity of the color present in front of it. The vehicle is installed with four DC geared-motors for driving the vehicle and two such motors for operating the flaps to capture the ball. The code for controlling the vehicle is written in C and the vehicle is connected to the computer through parallel port.

Automatic classification of Materials based on Microstructure Images

The system includes a module developed in MATLAB capable of recognizing materials based on various features derived from their microstructure images. The system captures images using a high resolution camera mounted on a microscope. Based on the image captured a series of features are evaluated. These features are then used to identify the material using nearest neighbor classifier with the already stored features in the database. (Developed in association with Sachin Agarwal (M.Tech student) at Motilal Nehru National Institute of Technology, Allahabad under the guidance of Professor Rakesh Mathur, Applied Mechanics Department.)

Digital Stop Watch

The major goal of this lab assignment was to develop a digital stop watch. The basic interfacing of LCD module, MAX7219 7-Segment display driver, speaker, shift register for capturing inputs using switches and pulse generator is first achieved. The successful implementation of the stop watch is possible with synergistic integration of the components in the system. A digital stop watch is developed which operates in two modes i.e. normal stop watch mode and timer mode. The timer mode provides the user an option to enter hours, minutes and seconds after which the watch should give a buzzer. Both the modes provide option to pause or continue and stop the watch. In addition, the watch also has the feature of Auto Sleep Mode which kicks in when the watch is idle for some time. The user can also turn this mode off. Also, the watch provides audio visual cues to the user for user friendliness.

Presentation       Project Report    Source Code


Digital Multimeter

The major goal of this lab assignment was to develop a digital multi-meter. The basic interfacing of LCD module, shift register 74X165, shift register 74x595, ADC, operational amplifier, relay, MOSFET is achieved with the BASIC Stamp II. A digital multi-meter is developed which operates in four modes i.e. Ohmmeter, Voltmeter, Digitally Controlled Voltage Source using R-2R, Digitally Controlled Voltage Source using PWM. All the modes also provide the feature to make multiple measurements and retain the last two readings. In addition to this the developed multi-meter also offers a feature to check the continuity of a circuit. Also, the system has the feature of Auto Sleep Mode which kicks in when the system is idle for some time. The multi-meter also provides visual cues to the user for user friendliness.

                  Presentation     Project Report     Source Code

Speed Control of a DC Motor

The goal of this assignment was to develop a Speed control system for a DC motor. Various control methods were to be employed, both open loop and closed loop. All control methodologies are implemented using MATLAB GUI. The important feature of the system is the auto-calibrate function which provides the capability to calibrate the system at three different levels depending on the desired level of accuracy and recommend whether to use the calibration or not based on the statistics of fit. In closed loop control, on-off, differential, proportional, proportional-plus-differential and proportional-plus-integral-plus-differential are implemented. The system also has a feature to run a diagnostics in order to assess its state of health. A monitoring feature which employs a webcam to monitor the fan by the user is also present. This can be of importance in remote operation of the motor. In addition, the system also provides the feature of data logging both for calibration and for controls.

Presentation     Project Report     Source Code

For any queries or suggestions please drop me an email at   or submit comments on Contact page.
hit counter Visits since June 11th, 2010
(Last updated on Mar 23rd, 2017)
Best Viewed in 1280x1024