Introduction to applications in Robotics

T A DwarakanathDivision of Remote Handling & RoboticsBARC 

Introduction to applications in Robotics

Probably Highest User of Robots in the Country

• The reason of course is the hazardous environment that exist in different stages of Nuclear Fuel Cycle.

• Inspection & Monitoring is very critical. • BARC has a mandate for developing technologies for societal application 

• Research for finding robotics substitute for manual applications. 

The Big Picture: Nuclear Fuel Cycle

The main Component of NFC: The Nuclear Reactor 

The Vastness of Nuclear Reactor under Construction 

Robotics in DAE

is what is known as theMaster‐Slave Manipulators

(over 90% of present day robotics in BARC is of this type)Why? : 3 points.

• Can be applied to variety of applications in different forms.

• Emerging as potential new generation sensing, manipulation and manufacturing solutions.

• There is an high excitement about this  mechanism owing to its potential application space.  It is complementary to serial mechanism.

• BARC has built capability in various areas of parallel mechanism based technology. 

Reasons for In‐parallel Mechanisms

Serial Vs. ParallelSerial Mechanism 

Arrangement: open loop mechanisms.

Manipulator singularities can be easily visualized.

Provide larger Workspace 

DKP has a unique solution and is simple. While the IKP can have multiple solution and is difficult to solve. 

One Kinematic chain support the end effector.

The payload to self weight ratio is much is small.  

Lower end‐effector precision in position and orientation. 

Parallel Mechanism Arrangement:  closed loop mechanisms. 

Manipulator singularities are hard to visualize.

Provide smaller Workspace.  

DKP has multiple solution and complex.While the IKP has a unique solution and simple to solve. 

Multiple Kinematic chains support theend effector.

The payload to self weight ratio is high.

Higher end‐effector precision in position and orientation.

1. Six axis Force‐Torque Sensor

2. Mirroring motion algorithms

3. Intermediary Telepresence

4. Parallel Mech. based MASTER‐SLAVE arrangement 

5. Research on Single Drive Multi‐DOF PSS based Parallel Mechanism

6. High precision alignment for synchrotron beams 

7. Surgical Coordinate Measuring Mechanism (SCMM).  

8. Robot based Neuro‐surgical suite 

Activities on Parallel Mechanism in BARC: 

Articulated Robot

Most Common Architecture (60 -70%)

Parallel Mechanism based Robots

High Precision 6 dofPositioning Mechanism

Master‐Slave Manipulators

Most Promising Architecture

Serial Vs. ParallelSerial Mechanism 

Arrangement: open loop mechanisms.

Manipulator singularities can be easily visualized.

Provide larger Workspace 

DKP has a unique solution and is simple. While the IKP can have multiple solution and is difficult to solve. 

One Kinematic chain support the end effector.

The payload to self weight ratio is much is small.  

Lower end‐effector precision in position and orientation. 

Parallel Mechanism Arrangement:  closed loop mechanisms. 

Manipulator singularities are hard to visualize.

Provide smaller Workspace.  

DKP has multiple solution and complex.While the IKP has a unique solution and simple to solve. 

Multiple Kinematic chains support theend effector.

The payload to self weight ratio is high.

Higher end‐effector precision in position and orientation.

The special interest on parallel mechanism is   because of its variety in architecture and its usefulness in its different forms 

1. As a rigid structure,2. As a complaint structure 3. As a passive device 4. As a controlled manipulator.Each form has uses in variety of applications.

PARALLEL MECHANISM

Technology Development serving in applications like 6‐axisCoordinate Measuring Machines, Parallel Mechanism basedMaster‐Slave Manipulators.

Master‐Slave Manipulators

Parallel Mechanism basedMaster‐Slave Arrangement 

T A Dwarakanath & Ketan Lagoo

Parallel Mechanism based Mechanical  MASTER‐SLAVE

Parallel Mechanism Devices (ZERO DOF)

Model of stiff wrench sensor

6‐axis stiff wrench sensor

{F}6X1=[J]6X6{f}6X1

The S‐S Chain 

Hexapod F‐T SensorDimensions:Height: 42 mmPlatform radius: 15mm

Set up for small force‐torque measurements

Specification

Force Diagrams : Graphical Force Perception

Graphical representation of wrench information is devised 

Force Diagrams, which are proposed here, offer meaningful Graphical representation of Forces and 

Moments.

Constructing Force Diagrams

• Consider work space model and wrench information

• Compose force diagrams from Forces and Torques so as to highlight the key features  

• Compare intended and actual force diagrams 

Force perception during 2‐D task

Effect of Disturbance 

Force perception for Intricate 2‐D Task 

Intricate caricature drawn by the robot 

Moment‐graph drawn using only sensed moment data

Force diagram for State of Remote Site

(a) Directly on Perspex sheet.

(b) A sheet of paper on top of Perspex.

(c) Two sheets of paper on top of a Perspex.

3D‐Force Diagram with Spatial parallel manipulator

3‐D Force diagram with random noise

Isometric View

Resolving 3‐D Force Diagram in 2‐Din Real Time

Intended and Actual Force Diagrams

Usefulness of Force Diagrams

Force diagram can enhance the operator’s knowledge about 

‐contact wrenches‐real time force trajectory‐accuracy of the motion and force trajectory‐state of the remote site in terms of wrench state

Jacobix: An Analog Computer for a 6‐axes Force‐Torque Sensor

What is an analog computer?

‐OPAMPs can be used to create a circuit that can perform a specific complex computation instantly.

‐The circuit can process analog signals from transducers without A/D convertors.  

‐We use it to multiply  a matrix (a Jacobian) with a  vector. We call it ‘Jacobix’!

Its an Alternative solution

Develop a circuit that makes the sensor self contained. A typical electronics circuit can be repaired by any technician with circuit schematic.  Use conventional IC available in labs.

Implementation of Matrix Multiplier

Compliant Parallel Mechanism Devices

A Complaint Tripod Device

Tripod: 3 point complaint contact sensor to sense geometrical information.

Elastic elements of the connector

Compact arrangement of     elastic elements

The SPS Chain 

Compliant Sensor model and the prototype

Serial and Parallel manipulators with F/T sensors

Planar Parallel manipulator

Spatial Parallel manipulator

Spatial Parallel manipulator

Serial  Robot

F/T sensor connected to work table

F/T sensor connected to work table

F/T sensor in the wrist

F/T sensor  in the wrist 

INTERMEDIARY TELEPRESENCE FORREMOTE MANIPULATION

Industrial Telepresence Systems

• Immersive  – Real time video and audio – Feeling of touch and force– Low‐latency  

Problems with Haptic Telepresence

• Depends on Contact perception  • Delayed perception  • Serves only  one user at a time• History of the wrench signal is lost• Expensive• Complex

The objective of the work is to develop a practical intermediary telepresence framework for remote Master – Slave manipulation

The focus of the work is to conduct a reliable remote manipulation with good “presence” at the remote site

Intermediary telepresence 

INTERMEDIARY TELEPRESENCE• A full‐fledged telepresence system is not warrantedfor many tasks.

• The requirement is often such that the manipulationhas to be conducted remotely with a operator in thecontrol loop.

• The operator in the loop should play a master orsupervisory role and be able to manipulate thetelemanipulator and also take safety or correctiveaction.

Framework:

Feed-back

INPUT

MASTER

Human Operator(MASTER)

Command Input through Joystick or Master Interface

Feedback: Graphics Display

Data processing, Graphics display of input commands

Feedback data processing

Input controller code

Feedback signal processing code

Programmable Multi-axes controller

PLANAR PARALLEL KINEMATIC ROBOT

SLAVE

SLAVEMANIPULATOR &ENVIRONMENT

MASTER SITE

SLAVE SITE

MASTER CONSOLE

MASTER CONSOLE

Single Master : : Multi Slave mode

Spatial Parallel Manipulator with high Repeatability  and  Rigidity

Repeatability ≈ 30µmPayload to Wt. Ratio ≈ 2.5

. 

Parallel Mechanism Based Devices

Picture:SPKM demonstrating  a high accuracy spatial insertion

A  PLANAR PARALLEL KINEMATIC 

MANIPULATOR

Task space trajectory planning among cooperating robots through mirror motions

• Many tasks cannot be accomplished with a singlerobot but can be accomplished easily with two or morerobots.

• If the task demands high dexterity simultaneouslyfrom multiple directions or action at regions which areseparated apart, two or more robots are needed.

• To share the work (load sharing) in accomplishing thecommon goal.

Need of Co-operation among Robots(Mirroring motion algorithms) 

Tracing about a Mirror Plane

To complete the given trace by preserving the directionand flipping the start and the end points of the robots1 and 2.

Sharing the trace by flipping start and end points(Hands on Steering wheel Motion)

Gripping the load securely while moving it is one of the common applications in the hot cell.

Every point in the task space of Robot 1 will have a mirror image in the task space of Robot 2.

Load Sharing while tracing

Two robots in cooperation through mirroring motions

Special Application through mirror motion with two robotsin Cooperation

• A single robot to realize an ultra slow uniform motion is not easy.• The motors exhibit by motors is jerky motions at very slow speeds.• Two cooperative robots are set to move near around most proficient speeds, yet the

relative motion between their end-effectors can be programmed to have ultra slowvelocity.

Basic scheme for the ultra slow motion The possible smallest trajectory traced with the help of ultra motion scheme

The concept of mirror motions simplifies the planning ofcooperation among robots.

Many tasks can be broken into subtasks wherein thecomplete task can be completed under mirror motionscheme

The robot operators who are skilled workers with littleexposure to robot programming can comprehend andhandle the various tasks.

Simplicity in comprehension would reduce the humanerror and thus enhance the safety.

Mirror Motion Algorithm

Single Drive Multi‐DOF PSS based Parallel Mechanism

Research on Single Drive Multi‐DOFPSS based Parallel Mechanism 

PSS‐Application

Common shaft because the actuator axis is invariant 

Single Actuator PSS in an Array arrangement

Robot based Neuro‐surgical suite 

NEUROSURGICAL  SUITE – BARC,  INDIA

A robot based frameless Stereotacticsystem is developed which hasaccuracy comparable to frame basedsystem and patient comfort levelcomparable to frameless system. Itremoves line of sight problem. Itautomates the frameless Stereotaxyusing a high precision robot (ParallelMechanism based Robot).

Supervisory control robotic system

Telesurgical robotic system

Shared control robotic system

CLASSIFICATIONOFROBOTIC SYSTEMS

ROBOTIC SURGERY(CARDIOVASCULAR)

Multilink Arm

Surgical serpentine robots (Catheters)

.

Brain Tumor

Neurosurgery 

Craniotomy Minimally Invasive Surgery/Stereotactic 

Surgery

Frame Based Stereotactic Surgery

Frameless Stereotactic Surgery

Camera Based Neuroregistration and Neuronavigation 

System(SCMM based)

radiotherapy chemotherapy

Non‐cancerous Cancerous

(A) Inside view of brain (B) CT Scan of Brain

(C) Craniotomy (D) Minimally Invasive Surgery

(E) Frame based StereotacticSurgery

(F) Camera Based StereotacticSurgery

ROBOTIC INNEUROSURGERY

YEAR  Neurosurgical Robot

SERIAL 

1990 to 2000 PUMA 200Minerva

2000 TO 2010

2010 TO ‐

NeuRobot ResearchGeorgetown robotSpine AssistNeuroArmNeuroMate ‐ COMMERCIAL 

PARALLELEvolution1  ‐ Res. PaperNEUROSURGICAL SUITE – BARC. 

World Scenario of Neurosurgical Robots  

Surgical Coordinate Measuring Mechanism (SCMM) 

A passive 4 DOF SCMM with position sensors mounted ateach joint to determine relation of surgical tool to fiducialframe. Portable, customized Coordinate MeasuringMechanism for neurosurgery.

Repeatability: 80 µm, Accuracy: 300 µm.

Neurosurgical Suite for RobotAssisted Neurosurgery • Patient data preparation from imaging.  • Development of SCMM• Neuro‐registration using SCMM• Neuro‐navigation using SCMM• Visualization for surgical aid. • Development of a high precision robot. • Integrating Visualization & Robot for    conductinghigh precision surgery • Perform frameless Stereotactic Neurosurgery

Hexapod based CT‐biopsy

‐10

0

10

20

30

40

50

60

70

‐150 ‐100 ‐50 0 50 100 150

Series1

‐10

0

10

20

30

40

50

60

70

‐60 ‐40 ‐20 0 20 40 60

Series1

Rotation about  Z

Rotation about  Y

‐10

0

10

20

30

40

50

60

70

‐40 ‐20 0 20 40

Series1

Rotation about  X

Robot based Neuronavigation and Stereotactic Neurosurgery

Robot base Neuronavigation and Stereotactic  Neurosurgery

Thank You 

ACKNOWLEDGEMENTS

Dr.  D N Badodkar, BARCMr. K Lagoo,Mr. M N Rao,Dr.  Gaurav BhutaniMr. S K Sinha