Naveen Kuppuswamy

Naveen Kuppuswamy

PhD Student

Phone: +41 44 635 2401

naveenoid@ifi.uzh.ch

Website

About Me

With a background in Robot Control, my primary interests are in sensori-motor control learning in embodied systems. In particular, I am interested in the dimensionality reduction techniques for motor control of redundant/compliant robot systems inspired by nature.
I am also interested in various aspects of robot mechanical and electronic design.

I am/have been a member of the following projects :

1. AMARSi : An EU FP7 Project investigating Adaptive Modular Architecture for Rich Motor Skills. In the context of the project, I shall be studying design principles for architectures consisting of adaptive modules considering morphological computation.

2. RobotDoc : An EU Marie Curie Network project, investigating the development of different sensorimotor and cognitive capabilities in humanoid robots. In the context of the project, I shall be studying a "Quantitative framework for sensory-motor strategies". In particular I am working on learning Movement Primitives, and on understanding and exploiting body dynamics and material properties to self-learn and optimise motor behaviour for humanoid robots.

3. Octopus: An EU FP7 Project which aims to investigate and understanding the principles that give rise to the octopus sensory-motor capabilities and incorporating them in new design technologies to build an embodied artefact, based broadly on the anatomy of the 8-arm body of an octopus, and with similar performance in water, in terms of dexterity, speed, control, flexibility, and applicability.

It might seem a pretty tall order considering the current state-of-the-art but keep watching this space!

Active Research

1.Motor Primitive inspired Reduced Order Control of Redundant Robots : Inspired by neuroscientific theories of biological motor primitives and spinal force fields, I am trying to propose a control architecture for redundant (and compliant) robots based on reduced-dimensional control. Utilising tools from Model Order Reduction, and ideas from developmental psychology, the aim is to have a self-learning architecture that is tightly coupled to the sensors, actuators and the dynamics of the robot body.

2.Curvature Dynamic Model for Continuum Robot arms : Models of continuum systems are difficult to develop and not straight forward to control. Inspired by the reduced dimensionality in the Peripheral Motor System of the Octopus, I am trying to build reduced order models of the behaviour using flexure (curvature) sensing.The aim is to then utilise such a reduced order model to simplify the control problem of octopus arm reaching.

3. DMP for Compliant Tendon Driven Systems : I am researching the learning of Dynamical Movement Primitives for compliant tendon driven robots, and in particular, the role played by body properties in the learning.

Shelved Research

1. Mag-E Actuation : We are trying to design an actuator for fin-based locomotion based on the principles of electromagnetism. This design will allow us to study energy efficient actuation using passive properties of the underlying magneto-mechanical system and develop simple control strategies exploiting the inherent nonlinearity of the system. This is a colloboration with Mr. Juan Pablo Carbajal. This project has now shifted to become a simulated study on this mode of actuation aimed at achieving energy optimality in swimming.

2. Passive Silicone octopus-like arms and String Pullers : The first target in the Octopus Project s to come up with an effective design for actuation of soft-bodied arms which can replicate the prototypical behaviours of the natural octopus like reaching and grasping. We are trying to understand the passive properties of silicone and how this can aid us in the design and actuation of the string pulling arms, which are basically tendon driven silicone structures with inextensible strings simulating muscle contraction in longitudinal and transverse directions. This is a colloboration with Mr. Matteo Cianchetti of the ARTS Lab of Scuola Superiore Sant'Anna, Italy. I am now working on model validation using a curvature based measure on the real silicone robot arms.

Academic Qualifications

2007 Master of Science (MS) in Electrical Engineering and Computer Science, KAIST, Daejeon, Korea, at the Robot Intelligence Technology Lab.
2005 Bachelor of Engineering (BE) in Instrumentation and Control Engineering, Anna University, Chennai, India

Industry Experience

2008 Yujin Robot Co. Ltd., Seoul, Korea (Engineer, R&D).

Personal

Hometown : Chennai, India