Good day, Dr. Deepak Kumar Yadav. I'm Arjun. I've had a look through your background and your professional record. I'm interested in hearing more about your work and how you see yourself fitting in here with our faculty. Ready to start?

So uh Deepak Kumar Yadav and I have graduation like Btech in electrical and electronics engineering.

Whole engineering department.

And after that work for Bahara University from for in the Department of Electrical and Communication Engineering as an assistant professor for almost like 18 to 19 months. Then I joined join as PhD scholar in IIT Kanpur under the Department of Mechanical Engineering where is a program called Nuclear Engineering and Technology.

Program.

And I work there on pressurized water reactors and.

Their control for load following operation where I use the nonlinear control and robust nonlinear control techniques for the load following operation in pressurized water reactors. After that I went for my postdoctorate in University of Toronto, Italy. There I work with.

Uh, tracked autonomous vehicles, and I developed the nonlinear strategy for the, uh, tracking control of the tracked autonomous vehicle. Uh, after that I came back to India and I'm currently I'm working with, uh, uh, Central Mechanical Research Institute, which is the, uh, which is the unit of CSIR in Durgapur.

Here I'm working with the manipulators, uh.

UH-22 projects first is IIT intelligent assist device and 2nd is, uh, Apple harvesting, uh, automate autonomous system for A and I'm, I'm doing the path planning, kinematic dynamic analysis design and, uh, their implementation in hardware and software also. So that is what.

Uh, till now. So please.

Thank you for sharing your detailed background, Dr. Yadav. Let us now assess your expertise in areas pertinent to the role of a Mechanical Professor. I’ll proceed by verifying specific key skills relevant to the position. Let's start with automotive systems. Can you explain the fundamental working principles of a hydraulic braking system in automobiles, and how energy conversion occurs during the braking process?

So actually my background is not, uh, in mechanical engineering, uh, my graduation and other things. So, uh, even though I know little bit about, uh, about it, but, uh, I cannot uh, umm, tell here because before.

The before conforming all those things, uh, I'm basically an electrical engineer, Electrical and electronics engineer working with the mechanical fields.

Where specific fields are robotics and other things. So for this I can say I don't know for now.

Mm-hmm.

Hello.

Mm-hmm.

Could you explain how you would structure a theory-based course on robotics for undergraduate students, ensuring a balance between foundational principles and applications?

OK, so.

Uh, hello.

Great. OK, OK.

OK, so.

Umm. I will design it like first first chapter. Like I will break it in into at least three to four units. First will be.

Uh, to teach uh, like uh, basic course on uh robotics for the undergraduate, uh, uh, students.

Yeah, yeah, yeah.

Take your

Hello so first.

Yeah. So first will be the, uh, uh, first will be mechanical systems, uh, they're, sorry, dynamical system, their properties and their development, their system design and all other things. Uh, second will be their, uh, like, uh, uh, their control and their kinematics and dynamics and other things. And final, we will, uh, go for some.

Real system designed and their implementation to our lab.

Whatever we have the facilities, their implementation to those systems.

Thank you, Professor. How would you ensure that students effectively connect theoretical concepts with practical applications during the controlled and kinematics module?

So, uh, it will be done by uh, the doing the all theory, uh, theoretical classes with the experiment.

Uh, uh, alongside of that.

Like if we are, uh, teaching them the kinematics or dynamics, we can also show the, if you have the, uh, any robotic or something, any other mechanical thing. So we can experimentally show them, uh, uh, the results of kinematics, uh, in, in the same lab environment. So that will. That how they will connect the theory and practical.

At the same time.

Yup.

How do you approach evaluating student performance in both theoretical and laboratory courses to ensure a fair and comprehensive assessment?

So for the theoretical courses, uh, I think we should assess them, uh, on the basis of how they have, uh, uh, how much they have learned the.

Basic, uh, principles of that course and that should be assessed by, uh, uh.

Like solving.

By class examination and other way where we we, we can create some questions or direct to assess the their understanding of basic principles and that will be the way and for the experimental like practical aspects we we can give them. Something to design or something to create? Something to solve some problems?

And that they can take time and solve that. So this both way we can check how they have learned or not.

MMM.

Now let's focus on guiding student projects and research. How do you mentor students in selecting and executing meaningful research topics or projects, particularly in robotics or automation?

So, uh, we should, for if we are talking about the undergraduate students, we should focus on the, uh, we should focus on the topics like.

Like with all the topics like which are more related to the basic principles and for the higher higher grade students like PG and PhDs we should take take them through.

So we should we should offer them.

Real life.

On the edge, uh, control aspects or uh, automation aspects and problems uh, to be solved.

To be solved for their projects. So this will be the way I I think we should go.

Thank you, Professor. That distinction between foundational projects for undergraduates and advanced, real-world problem-solving for postgraduates is a clear and structured approach. Moving on, how do you ensure clarity and engagement when explaining complex mechanical or robotics concepts to a diverse group of students?

So I think it needs some preparation from my side also that I understand the which type of crowd I have to whom I have to deliver my lecture and what are their understandings are the way they learn actually. So if the people are advanced in mathematics and other things and they can learn.

That's, uh, mathematical, uh, systems easily. So I will go through initially with the mathematical system, uh, mathematical explanation of the system that actually clears, uh, the system.

Very makes it makes the system very intuitive. And if the crowd is not much in maths, mathematical analysis, uh, I should go initially with the some experimental setups.

And try to approach them to try to create their curiosity for the. Then I will, uh, go through some basic mathematics. And then after I can, uh, go for that complex idea which I want to explain.

Thank you, Professor. Assessing the audience’s needs and adapting your teaching approach is indeed essential for effective learning. Let's proceed to research. Could you discuss one of your research publications, preferably in robotics or a related area, and explain its significance to the field?

So actually in my masters I have worked with the nuclear reactors and their control where I have worked with.

Sorry where I have worked with the.

Modeling, mathematical modeling and their simulation of nuclear reactor core with their thermal hydraulics and neutronics and their control design. So I have published 3 papers on that and after after that I have worked with the robotics. So there I have developed a path planning algorithm. For the Apple picking application.

Which is a midpoint stretch path planning technique which is already.

Accepted and is going to be published very soon. So this is what I I have done till now in the field of automation, control and robotics.

Thank you, Professor. Both your reactor modeling work and your upcoming publication on path planning for apple picking demonstrate strong research contributions in automation and control. On that note, how do you see your research evolving in the next five years, particularly in robotics or automation? What areas do you aim to explore further?

Further, I want to explore.

Some some application area, some yeah like soft robotics and robotics for critical environments like nuclear reactors where the robotics is the main thing which which is used to which can be used to reduce the harmful effect to the human like. Whenever we have to go to the the reactors for inspection or some.

Some.

Like some maintenance work or other things, we can use the mobile robotics, mobile robots with the polluters or mounted on them. So we can even we can also.

Use the like sensors and other things to monitor and also to solve some to do some maintenance techniques there maintenance to solve the maintenance queries there. So this is the open area where I can do good work but even even. If it is not possible anyway because the the.

Uh, because the don't have the critical scenario available all the time. So we can also, uh, uh, go for another, uh, research areas with the same, uh, uh, created algorithms and devices.

Thank you, Professor. Your focus on robotics for critical environments like nuclear reactors highlights a commendable effort to address significant safety concerns through advanced automation. Exploring soft robotics in this domain also shows forward-thinking. Next, let’s delve into your experience with industry projects or consultancy. Could you describe your involvement in any such projects and your specific contributions?

So actually, uh, I have MMM, whatever I am doing now.

Uh, uh, uh, it's a project which is, uh, which is, uh, which is.

Which is industry project only uh, it is developed for developed. We are developing it for industry only. So one is IAD, it's a intelligent assist device and second one is uh, Apple picking project in which.

One is we are developing for the.

Any industrial process where we can.

Crane can pick up the load, place some other place, place at other position that is very heavy. The crane can lift a very heavy but it's not completely automated. But human can guide that load and. The the guiding guiding portion will be.

Guiding person will be human but the taking care of load and other things will be machine and 2nd is the.

Which is for the agriculture purpose. The apple picking automation system where the.

Robotic hand will pick.

The apples and, uh, collect the apples and also it will be able to identify which apple is, uh, able to pick or which apples we should not have to pick. And so these are actually the industrial projects.

Thank you, Professor. Both projects—intelligent assist devices and the apple-picking automation system—clearly demonstrate impactful applications of robotics in industry and agriculture. Your focus on human-robot interaction and automation for specialized tasks is particularly noteworthy. Finally, how do you align your research and industrial experience with mentoring and contributing to academic programs as a professor?

So, uh, with my research and, uh, industrial experience, uh, I can think, uh, that knowledge, whatever I have got the, the, uh, way I have, uh, reached the, all those, I have overcome the, uh, obstacles when I, it came.

Uh, those with me and, uh, by this it's, I know, I understand that, uh, student, uh, students should learn some practical aspect also. So I should provide them, uh, uh, I will provide them some, uh, practical, uh, uh, experimental systems to learn from, uh, real things. So that I think I have learned and what I can provide them.

Away I can share those learning to my students, and the students can learn.

Thank you, Professor. Your emphasis on translating your real-world experiences into practical, experimental learning for students reflects a strong commitment to both teaching and mentorship. Let me recap and finalize this segment.

Professor Yadav, let's begin. Could you describe which particular introductory course or fundamental concept in mechanical engineering you find most rewarding to teach?

So I can teach UH dynamics and UH dynamics and control of uh linear systems or nonlinear systems and also the.

Control system engineering for the undergraduates and automation systems.

That.

Yep.

I see that you have an impressive background, particularly with your work on agricultural robotics for apple harvesting and the advanced control schemes you developed for nuclear reactor operations. Let’s discuss the agricultural robotics project. Can you explain the novel path-planning algorithm you developed for the 7-DOF arm? How does it address the challenges specific to agricultural robotics?

So in agriculture robotics, the our project actually focuses on the apple picking where we we are targeting a arcade like structure where apples are there and in that case the apples are mostly.

Mostly in front of the picking a picking side so, but sometimes there are some branches and other things which obstacle between.

Which is the obstacle between robot arm and apple. So we we were we need a algorithm path planning algorithm that can that can create a path between the factor to apples and then it will pick the apple and put it to the basket. So we need a path planner which which can execute it faster. Which can execute this, you know faster way.

And SIM simple so we can make it fast like.

It could not. It should not be.

Like.

The sampling based because sampling based they have very heavy.

Computation.

But simple but good success, right? So we we developed a deterministic path planning algorithm which is basically based on.

Cost and all. So we wanted it simple and.

Uh, detour technique, uh, in which we actually, uh, create not between, uh, .1 to .2 where from, uh, we have to create a path like from home position to Apple. And if any obstacle comes in between, we check for that and we stress the midpoint of that path.

To, to, to the far, to that obstacle, and.

If it's again, uh, in obstacle, any portion of line, we again follow this, uh, this method to stress the midpoint of that line. So this recursive, uh, deterministic planner, uh, finally we get a path which is not in collision. So this is the basic idea of the path planner. Some other, uh, improvements had been made.

In it, but this is the word basic. Basically this planner does so.

Finally we get a path which is not colliding and as it is deterministic and very easily solved, so it's fast and.

Manipulators. So by simulation we have checked that.

Too high, like 2192%.

Uh, that is too good for uh uh, for such a new deterministic, uh, path learning approach.

Your approach to using a deterministic path-planning algorithm for apple harvesting is innovative, particularly with its focus on simplicity and speed over computationally heavy sampling-based methods. It seems highly applicable given the constraints and obstacles present during agricultural operations. How do you see this algorithm being integrated into a real-world agricultural environment? For example, how would you handle variables like moving branches or dynamic obstacles in such settings?

So presently it is designed for aesthetic environment, doesn't consider any dynamic.

Change in the environment, but for the but the further dynamic thing, dynamic obstacles, we are going to develop a new assessment system which actually. Identity dynamic.

Portion of the picture and like if branch is moving at all. So that is the problem with identification when we will identify a dynamics.

Branch or dynamics dynamic.

System, uh, we will integrate this with this planner and then we will see how it works. So it, it, it is in process actually.

Understood. It’s a thoughtful progression—starting with static environments and moving toward dynamic obstacle handling with enhanced sensing systems. Let’s transition this to an educational perspective. How would you teach undergraduate students about implementing such a path-planning algorithm, considering they may not have extensive prior knowledge of robotics or computational methods?

So give him some.

Move to daylight problems like if.

If they want to go from one point to other point, they should know.

The paths from which they can go without getting.

Getting much traffic or something. So how they decide the things can be done? So that is the way.

By practical examples we can go, we can teach them. So we can take multiple examples like.

Day-to-day 1.2 Another point, transition another multiple things.

So I have to think about it that and I effectively tell them. How can I effectively approach these ideas to, uh, teach them?

Time of, uh, lecture. I can uh, I think uh, it needs some preparation to.

Uh, at the.

Think about it.

Your approach of connecting complex algorithms to real-world scenarios, such as navigating through traffic, is an excellent start for introducing students to these concepts. These relatable examples help in building intuition before diving into technical details. Taking time to refine those lecture materials shows your commitment to effective teaching. Let's turn to your other project on industrial robotics and the design of admittance and impedance control algorithms for human–robot interaction. Can you elaborate on how you addressed the primary challenges in ensuring safety during physical interaction with humans?

Uh, sorry, I, I didn't follow you. I, I want to, uh, hear it again, please.

Of course, Professor Yadav. In your project on industrial robotics, where you designed admittance and impedance control algorithms for human–robot interaction, could you explain how you addressed the key challenges to ensure safety during physical interactions with humans?

Yes. So actually impedance and arbitrage control are used for the safety plane where the industrial system should not go beyond a safety plane if you have already defined a safety plane.

Around the industrial setup, the industrial arm should not go through that safety plan to the other side of safety plan that will actually make the make the area outside of that safety plane.

To the people and this is uh, implemented the impedance and uh, admittance control has been implemented at that boundary point uh, like if the robot arm goes beyond that boundary point, uh, force will uh tell it to come back and when it is in the area of their working area it it will be. Follow the how?

Human will, uh, need to follow like, uh, if uh, we direct.

By hand, uh, in this, this direction, uh, in which direction we want to direct them, it will follow that direction and uh, in which force and in which velocity we want to direct them, it will follow that velocity. So this is how we are, uh, using the safety.