konnect Issue-date: 30th September 2020.

AUTOMATIC CONTROLS AND ROBOTICS DIVISION


Automatic Control and Robotics : Quick Journey
Dr Jayesh Barve, Past-President, ISA Bangalore Section (Work: GE Research, Bangalore)
barve.jayesh@gmail.com
(Courtesy: ISA-ACARD Newsletter Jun’20 to allow republish here.)

A) HUMAN BEHAVIOUR & PERSPECTIVES
Human Behavior: It is analyzed and proven biologically and psychologically, that humans always have an open (or hidden) tendency, desire or fantasy (not only in today’s era, but for ages since the human existence); to find or have “someone else” do tasks that are (or becomes) routine or unpleasant. For centuries, as is evident in epics & history, that humans with power &/or wealth, even tend to fulfill their such desires through (legal, illegal or sometime even inhuman) Animal, “Human-Slaves” or “Human-Servants” -- as EITHER forced labour/prisoner/slaves OR employed workers -- to help them in daily work, and/or to help fight for them during wars. On the other side, human is sensitive and kind. So, also looks for options to correct, improve such inhuman acts (of having kept/slavery/servants) by replacing “human-slaves/servants” by “non-human slaves/servants”; not only in science-fictions & story books, but also in real-life. Such replacements were earlier in the form of providing some aids/tools/dummies to “human-slaves/servants” to reduce sufferings and hazards during work, wars. Later, such aids/tools/dummies started moving up the ladder to make lives of such “human-servants” better by improving their work quality & efficiency. Today, known as “Automatic Control”, “Automation” or “Robots”.
The term “Robot” was probably 1st time coined by Czech play writer Karel Capek in 1921 science-fiction Czech play “R.U.R” (“Rossum's Universal Robots”). Robota (Czech) means “A worker of forced labor”. This play begins in a factory that makes artificial people called robots, human-like creatures, like “androids”, “humanoids” we call today. These are artificial beings designed to resemble like a human, made by human, to help doing tasks previously/otherwise done by human either for day-to-day works or to fight in wars.

Ancient Mentions: RigVeda (old Indian book ~1500 BC) mentions “Asura” as the one who has magical or phenomenal powers. One such Asura namely Sambarasura, used his powers to create 3 human-like slaves (robots) named dama,

Fig.1: Human-aids - flying chariot, bow-arrows

vyala and kaTa. These 3 Robots were lifeless machines and so had no sentiments/emotions, hence were never defeated. Also, in another old Indian philosophical book “Yoga Vasishtha” (a discourse of sage Vasishtha to Prince Rama, authored by sage Valmiki centuries BC) mentions Robots & Artificial Intelligence (Maya). Also, another Indian epic book “Ramayana” (~500 BC), describes things like many software (Mantra) powered bow-arrows as war-aids, & flying chariot (Pushpak Viman) for fast mobility.

Automatic Control & Automation represents something that is created by human to perform the tasks previously or otherwise performed by humans; but with a reduced, minimal or no human-intervention. It is Labor-Aiding, Labor-Saving means to perform process or procedure with reduced/minimal human assistance.

B) AUTOMATION – OLD HISTORY:

Fig.2: Heron’s Automatic Door & Aiolipile

Heron can be named as one of the great ancient “automation engineer”, who devised and documented many automatic devices (~10-50AD in Greece) including famous automatic temple-doors using heat, pneumatics & mechanics; and Aiolipile –1st steam-powered engine (Fig.2).
Also, Greeks and Arabs during ~300 BC-1200 AD were fascinated to invent ways to keep accurate track of time. In Egypt ~270 BC, Ctesibius described a float regulator for a water clock (Fig.3), an earliest feedback controlled system. The Persian brothers Banu-Musa, in their “Book of Ingenious Devices” (850 AD), described number of automatic controls e.g. 2-step fluid level control, form of discontinuous variable structure controls. They also described a feedback controller.

Fig.3: Ctesibius Feedback-control Water-clock

But, the Fly-ball governor speed-control mechanism (Fig.4) invented by Christian Huygens & Bunce (separately for different applications) and used by James Watt in his famous steam-engine, can be considered as 1st feedback proportional controller of the revolutionary 18-19th century industrial era.

Fig.4: Flyball speed-governor by James Watt

In 18th century, Swiss mechanician Henri Maillardet built an interesting non-industrial Automatic machine "Draughtsman Writer" (Fig.5), is believed to be the one with largest mechanical "memory" (4 drawings, 3 poems) of any such mechanical automatic machine ever constructed.

Fig.5: The draughtsman-writer Automaton

C) AUTOMATIC CONTROL – IN CENTURY:
In the 1st half of 20th Century AD, advances in electrical, electronics transformed traditional mechanical ON-OFF & proportional-controllers to electronic proportional-controllers e.g. amplifiers for better noise control by negative feedback.

Automatic Control Theory: Also, around 1st world-war, James Maxwell through his work established theoretical basis of control theory, which was then continued further advancements by other engineers like Evans, Bode, Nyquist, Wiener, Kalman et al. Alongside, proportional controllers got extended to marvelous PID (Proportional Integral Derivative) controllers of mid 20th century, which are the most-popular and widest used controllers, even today.
Around 2nd world-war, a German woman engineer Professor Irmgard Flügge-Lotz introduced theory of discontinuous automatic control leading to theoretical treatment, developments & applications like ON-OFF hysteresis control, aerospace guidance, and formed basis of non-linear controls.

Automatic Control Platforms: On the other side, after suitable control theory background laid-down, along with engineering & digital computing advancements, ON/OFF & PID controllers got implemented in various forms like Analog hydraulic-pneumatic-electronic; Digital microprocessor based computerised ON-OFF, Discrete PID, Sequence controllers.
Further, in 2nd half of 20th century, with the fast-growth & miniaturization of electronics, instrumentation, analog/digital communications, microprocessors, computers, and software; programmable, flexible, rigorous & complex mathematical forms of automatic controls became possible & viable (otherwise impossible or cumbersome just using mechanical/ pneumatic/analog-electronics controllers). Hence came into existence the Digital/Embedded control-systems (with programmable, simple & complex digital controls); PLCs (flexible programmable logic & sequence controls), DCS, SCADA (plant-wide controls). Alongside, control theory advancements journey via state-space, adaptive, optimal, model-predictive, robust & stochastic control schemes & algorithms etc. led to more feasible, viable real-life industrial & non-industrial controls applications. These automatic control advancements helped provide better performance, efficiency, reliability, robustness, flexibility, programmability.

Automatic Control Applications: With success & popularity of PID controllers, more & more applications enabled automation i.e. aid/replace humans, partially or fully, in various devices, equipment, subsystems, systems, unit and plant-wide operations in – 1) continuous manufacturing industry (like chemical, petrochemical, oil & gas, paper & pulp, mining, power-generation, etc.); 2) discrete manufacturing industry (like automotive, pharmaceutical, food-processing, dairy etc.); 3) non-industrial/hybrid systems/equipment/devices (like home-appliances, consumer goods, automobiles, locomotives, aircrafts, packaging, machinery, medical equipment, etc.). Without advances in automatic control, the most operations in such industry/non industry applications would have been manual; making life of humans miserable and hazardous OR several applications would have been even impossible.

Recent & Future trends: Over last ~2 decades, with the availability of viable, powerful industrial computers, it has become possible to take automatic control beyond traditional single/multi-loop ON-OFF/PID controllers (deployed at field control, PLC, DCS levels). Rather, the use of automation has penetrated now even at upper hierarchical layers of industrial plant-operations (Fig.6) like process control & optimization layer; production planning & scheduling layer;

Fig.6: Advanced Hierarchical Control in Industry

& even multiplant coordination covering supply-chain management layer. This has become possible due to automatic control theory, algorithms & computing accomplishments like multivariable model-based predictive & (mathematically) optimizing controllers (e.g. MPC). Again, these led to allow automation to help/reduce/replace human tasks & interventions at upper, decision making layers. These advanced hierarchical automations resulted in benefits of enhanced/optimal operational efficiency, productivity & economics in industry.

Future: Even, these traditional and advanced hierarchical automatic controls have still some gaps constraining the reach of automation to many other applications like fast industrial systems & operations (e.g. power grid, aircrafts). But many such operations are possible & are performed by human. Because, humankind have a special capability of vision sensing (images/video) and cognitive decision-making. However, with latest ongoing research & advancements in sensing, microelectronics, communication, advanced powerful microprocessors/GPUs, and computing etc. have led to new interdisciplinary areas like IIOT, Edge-Fog-Cloud computing, distributed computing etc. These technologies can address these gaps and offer advanced solutions for such large-scale, fast multivariable systems. Additionally, increasing capability of intense image/video capturing & processing, artificial intelligence & learning algorithms (found promising in image/video information processing), have made it possible to develop & apply vision-based intelligent & cognitive decision making (controls) in such applications. These are sure to further take the goals & objectives of automatic control (automation) to help/reduce/replace human interventions in many remaining, laborious routine tasks e.g. monitoring, inspection, mobility, healthcare, etc. This automation journey is ongoing & will continue in future, due to fast evolving research to expand systems theory, controls and automation towards many new non-industrial applications like robotics, smart manufacturing, mobility, video surveillance, navigation & control, traffic management, home/building automation, grid power & energy management, logistics, supply chain, retails, mining, biological & life-science, biomedical (e.g. ventilators, pacemakers), waste management, business processes automation, social behaviour, epidemic/pandemic community dieses control.

D) ROBOTICS – IN CENTURY: As per Robot Institute of America, a robot is defined as a reprogrammable, multifunctional manipulator designed to move material, parts, tools, or specialized devices through variable programmed motions for the performance of a variety of tasks. Japan and Germany have been in the forefront of Robotics development, application revolution in 20th century. Japanese Industrial Robot Association (JIRA) simply defines Robot as “A device with degrees of freedom that can be controlled.”. JIRA classifies robots as -- Class-1: Manual handling device; Class-2: Fixed sequence robot; Class-3: Variable sequence robot; Class-4: Playback robot; Class-5: Numerical control robot; and Class-6: Intelligent robots.

Industrial Automation and Robotics revolution started along 2nd world-war and then grew fast. Various tasks where Robots have helped reduce/replace human tasks for betterment of humans are listed below. Though not exhaustive, this gives a quick glimpse to breadth of fascinating big world of Robotics:











a) Dangerous/Difficult for humans

  • Space explorations
  • Chemical spill cleanup
  • Disarming bombs
  • Disaster cleanup
  • Nuclear waste handling
  • Subsea Manufacturing/Assembly

b) Boring and/or very repetitive

  • Welding car frames
  • Part pick and place
  • Manufacturing parts

c) High precision or high speed

  • Electronics testing
  • Surgery, Precision machining

d) Specific Task Machines

  • Bottling machine
  • Dishwasher, Paint sprayer

e) Multi-task Machines

  • Pick & Place Arms
  • CNC Machines
  • Mobile Robots

f) Advanced Robots

  • Mobility, Human Interfaces
  • Haptics – tactile sensing
  • Snake, Insect robots
  • Humanoids, Soft-robots
  • Aerials & Underwater AVs

g) Non-engineering Applications:

  • BOTs in eCommerce, e-Business, e-Gov, PA Services We will talk more about Robotics in future newsletter.


Author’s Profile:



Dr. Jayesh J. Barve, PhD (Systems & Controls) from IIT Bombay is Principal (Controls & Optimization) at GE Research - Bangalore, India. He has 29-year experience ~15 in Industrial R&D and ~14 in Academia. He has published >15 patents and >45 papers in Journals/Conferences and >25 reports/whitepapers; contributed to ~8 next-gen products/technologies. His experience & interests spans across Controls, Optimization, IIOT, Automation, Autonomous/Guided Vehicles and Power, Energy, Microgrid, Oil & Gas, Appliances, Manufacturing, etc. He is a senior member with leadership roles & responsibilities in voluntary organizations like IEEE, ISA, and IFAC. He received several awards & certifications in GE Research. Recently, he received 2019 ISA “Excellence in Technical Innovation” Award endowed by Honeywell UOP company for technology R&D leadership contributions. He is past-president, ISA Bangalore Section, & a Conference Program Chair ISA ACARD.