This IIT Madras team is building rovers & drones for MARS EXPLORATION | BP2B: Student Edition! Ep.05

1. 0:00 – 0:48

   Intro

   1. ADAdithi0:00

      during the mission, we tried to climb up a 70-degree slope, and the rover ended up toppling over, and it fell, like, right on its back on the antenna. But when we turned the rover back onto its wheels, everything was completely fine.

   2. SPSpeaker0:13

      There's almost $50 billion worth of material up there, and you don't want to accidentally drive it into a ditch.

   3. SPSpeaker0:19

      We have built a 3D-printed in-house spectrometer. So using that spectrometer, what we do is that we search for signs of life, like protein or carbohydrate, in the soil.

   4. SPSpeaker0:28

      Found out that there was one bolt which was very loose. It just came out and was causing resistance to the rotation of the wheel. What the team learned was how small things actually matter. One screw not tight and, or one wiring going wrong could actually, you know, cost you a lot of things. [upbeat music]

2. 0:48 – 1:58

   Anveshak: Team Lead Introduction and what Anveshak does

   1. SPSpeaker1:07

      Hi, and welcome to Best Place To Build: Student Edition. I'm Vidhi, a fifth-year engineering design student at IIT Madras, and we're here today with IIT Madras' very own space robotics team, Team Anveshak. If you've ever watched The Martian and wondered how close are we to technology like that, you're in the right space. To learn more, I'm here with their team lead, Adithi. Hi, Adithi. Which year and branch are you from?

   2. ADAdithi1:30

      Hi, Vidhi. I am, uh, Adithi. I'm from fourth-year BTech Mechanical Engineering.

   3. SPSpeaker1:34

      So Adithi, could you give me a little idea on what Anveshak does? Space robotics is a very broad domain. What are you guys focused on?

   4. ADAdithi1:41

      Okay, so, uh, as Team Anveshak, we make prototype Mars rovers, and we have also introduced our very own drone module to complement the rover's, uh, operation. So, uh, what we do is we build robust, uh, land rovers, where, uh, we are essentially

3. 1:58 – 2:40

   Rover and Drone Partnership: How does it help?

   1. ADAdithi1:58

      trying to achieve the same level of operations as the, uh, well-known Perseverance and Curiosity Mars rovers that's made by NASA, and also similar to the missions like Chandrayaan 3, that was recently start- done by ISRO. So, uh, our main goal is to, uh, basically help out in these kind of explorations, build on, uh, similar space technologies and, uh, yeah, just overall fuel that, um, curiosity of whether there is extraterrestrial life out there.

   2. SPSpeaker2:31

      Great! And how does a rover and a drone partnership here help us in knowing more about extraterrestrial terrains and life on other planets?

4. 2:40 – 6:25

   Mars: Why Mars and how is the Mars environment replicated for testing?

   1. ADAdithi2:41

      Okay, so, uh, the rover in general can reach into places that generally a human astronaut cannot reach into. So the rover in itself is a very big, I, I could say, boon for space exploration, and to complement that with drone is to reach the parts where even the, uh, rovers cannot reach into. So to be able to, uh, bring about that connection between rovers and drones, that would, in my opinion, just expand, like, vast regions into space exploration.

   2. SPSpeaker3:11

      That's very interesting. And you mentioned Mars specifically, and I've heard of Anveshak earlier being known as the Mars Rover Team. So what is it about Mars that makes it so interesting?

   3. ADAdithi3:22

      Okay. So, uh, from the previous reach- researches that have been going on, it can be seen that the atmosphere that has been found on Mars is kind of similar to what we have on Earth. So it is more like a mission of finding of... finding out whether or not we could eventually, uh, habituate into Mars, the missions that have been conducted by NASA. So it is to, along with, uh, figuring out whether there have been extraterrestrial life, it is also about finding out whether, uh, there have been conditions on Mars before that were similar to Earth, or it can be developed similar to that.

   4. SPSpeaker3:59

      That's great. And right now, your rover, what all is it capable of doing?

   5. ADAdithi4:04

      Yeah. So our rover is actually a semi-autonomous rover, so which can also be, uh, remotely controlled, and it can also operate autonomously. So, uh, the rover has, uh, apart from its traversal capabilities, also has a manipulator, which can basically do, um, basic manipulation, uh, operations, like, uh, flipping switches, or picking up rocks, or, uh, taking in soil samples, and all that. And, uh, since space exploration, that cannot be very much... It's not reliable to, uh, go more on the remote control base. We also have an autonomous, uh, part on the rover, where it can basically autonomously navigate and also help out more in the space exploration.

   6. SPSpeaker4:55

      And how many people do you have on your team on board helping you achieve this?

   7. ADAdithi4:59

      So we have about 50, uh, students from, uh, various different engineering domains, uh, who are, uh, divided into five modules. So, uh, which is mechanical module, which basically works on the framework of the rover; the electronics and software module, which makes sure that the, uh, telemetric communication and autonomous delivery happens; the astrobiology module, which is the hype of the team, as you could say, because that is what, um, you know, focuses on the extraterrestrial life detection. And we have our very own drone module, and finally, we have the CPR module, which takes care of the finance, sponsorship, and everything of the team.

   8. SPSpeaker5:36

      Inherently, since you're a competition team, I am very curious to know how they are replicating or trying to simulate the environment on Mars for you to test here on Earth.

   9. ADAdithi5:47

      So, uh, as a competition team, we usually go to a few, uh, national and international competitions. So how that works is that they usually set up a terrain which is very much similar to Mars, so that's like rugged terrain. Then there is a lot of dust, craters, and everything. And, uh, so how it works is we also simulate a similar, uh, environment, that we have a separate, uh-... part called a base station, where people who are operating the rover when it is remotely controlled are not allowed to look at the rover, and they only use camera feed and stuff to, uh, basically navigate and, um, like, look

5. 6:25 – 9:22

   Competition Experience: Anxiety inducing or Exciting?

   1. ADAdithi6:25

      for particular places to explore. And, uh, these competitions, they're also, uh, systematically organized, in which we have like three types of missions, which would include, like, manipulation, as I had mentioned before, autonomous, and also astrobiology mission, where it is basically something that would generally be expected as a part of ex- uh, space exploration.

   2. SPSpeaker6:52

      Observing a rover only through some camera and sensor data, and not being able to see it in person during a competition must be very anxiety [chuckles] inducing. How has your experience been, uh-

   3. ADAdithi7:04

      Yeah

   4. SPSpeaker7:04

      ... when you've gone for a competition like this?

   5. ADAdithi7:06

      Yeah. So, uh, as I mentioned, there is a base station part of it, but there's also another safety runners that we have ba- which are basically like two people that are moving along with the rover, just to make sure that it does not crash or stuff. So I've been... Personally, I've been a, a runner. So, uh, being a runner, you do not understand what is going on base station, so you just are around the rover, and you think that the operator is doing just some random stuff, and they're trying to destroy the rover or [chuckles] something like that. So, uh, the experience has been surreal because, uh, as a person who is able to see something that I have built, I've worked upon something, and seeing that being tested under certain constraints, that was just so exciting and satisfactory that I would say... And yeah, there were a lot of hardships that have gone before. Uh, like for example, in one of the missions, like we- our rover is usually capable of, uh, climbing up slopes up to 45 to 50 degrees. But we try to... During the mission, we try to climb up a 70-degree slope, and the rover ended up toppling over, and it fell, like, right on its back on the antenna. But the most, um, I think exciting thing that happened was that when we turned the rover back onto its wheels, everything was completely fine, and we were able to quickly resume the mission and, like, also finish the mission. So that, I think, is an experience that cannot really be put into words. That excitement, that satisfaction is re- really surreal.

   6. SPSpeaker8:42

      So right now, it's post the exam season. Our end semesters are done. Placements are also done. Yet, I see this Anveshak team, like, pretty packed up over here. What are you guys doing right now?

   7. ADAdithi8:55

      So we are actually working towards a competition that is coming, uh, towards the end of January. It is called International Rover Challenge, which is something that happens here in India. So, uh, what we are working towards is, this is our 10th iteration of our rover, which is called, uh, Isaac. So funny thing about our team is that we, uh, alphabetically name our rovers. So we had two iterations in our, uh, in the H alphabet, and this is our 10th

6. 9:22 – 12:18

   Meet Isaac: Anveshak’s current rover and upcoming competitions

   1. ADAdithi9:22

      iteration called Isaac. So what we do in this, uh, like, one month up... leading up to the competition, is that we consistently test out our, uh, rover, and to check if there are any faults, to see different kinds of fields where we could improve, and also basically simulate the competition itself in order to like, um, get better at what we're doing. So we have spent sleepless nights here, so trying to make things work, and, uh, yeah, so we are excited about the competition.

   2. SPSpeaker9:56

      It's great to know. And how has Isaac been doing? As the team lead, what do you feel about Isaac's chances at winning the competition?

   3. ADAdithi10:03

      So from where we had started, I would say we are at a pretty good position, but I would also say that there's a lot more scope for us to improve along with the rover. So, uh, we have been testing day and night, and, uh, something that we would want for it to... is to win competitions, basically. So right now, I would say I'm pretty confident, with a scope of improvement.

   4. SPSpeaker10:25

      Of course, of course. Makes sense. And this competition would be at the starting of 2026, in January. Post that, do you have any other competitions lined up?

   5. ADAdithi10:33

      Yeah. So other competitions that we usually go to is, uh, something called IROC, which is ISRO Robotics Challenge, which is something that was introduced two years ago by ISRO. So we started, uh, the first year of ISRO was about, uh, developing a fully autonomous rover, and just last year it was about, like, developing drone that would basically complement the, uh, operation of a rover. So that is where we get our direct link towards the actual space exploration, because we are directly, you know, collaborating with ISRO on this. And, uh, similarly, we also go into another, uh, competition called IRDC, which is International Rover Design Challenge. It is an online challenge where we basically design rovers that are meant to hypothetically go to the, uh, Martian or lunar, uh, environment, and basically to test its resilience there. So these competitions that are designed, they are basically for us to get a glimpse of what space exploration would be like. So apart from this, we are also looking forward to taking part in University Rover Challenge, which happens in Utah, the Mars Desert Research Station, where is exactly where NASA rovers have been tested. So getting all of these brilliant opportunities to, you know, indulge into space technology is something that we do.

   6. SPSpeaker12:01

      That sounds like an incredibly factious, but exciting road at the same time. Like, testing on the same grounds as NASA, that's almost unbelievable. Uh, that brings me to the question: How has Anveshak performed as a team in the past competitions that you've had?

   7. ADAdithi12:16

      ... So some of the achievements

7. 12:18 – 14:22

   Anveshak Results: Competition wins and Achievements

   1. ADAdithi12:18

      that I would like to talk about here is that we came first in Caterpillar Autonomy Challenge, uh, this year, uh, in January, which is a part of the, uh, Shaastra that happens here. So, uh, we were also awarded for the Best Rover Design, and we also, uh, obtained a cash pool of 1.55 lakhs. And apart from that, we also came second in IRC 2024, and, uh, we came fourth in IROC, uh, 2024, where we had to design our very own autonomous rover. And, uh, this year also, we were, um, fortunate enough to be able to participate in a drone version of the IROC. Uh, we have also, uh, secured 12th position in URC, uh, 2019. Apart from this, we've also, uh, gotten the opportunity to represent IIT Madras in, on a, uh, national level, where we were invited for, uh, a preliminary testing on the Indian version of the Mars Desert Research Station, which was organized by Akka Studio, uh, in collaboration with PRL.

   2. SPSpeaker13:31

      That sounds like a very stacked list of [chuckles] achievements. Congratulations to your team. And it makes me even more curious how you have so many subsystems working together to make this rover function at the end of the day.

   3. ADAdithi13:44

      So to know more about the technical aspects of what goes into the mecha- mechanical module, so, uh, I have our mechanical module lead here, Ayush, who can tell you more about this.

   4. SPSpeaker13:54

      Hi, Ayush.

   5. SPSpeaker13:55

      Hi, Vini.

   6. SPSpeaker13:55

      Uh, thank you for taking the time. Uh, which year and branch are you from?

   7. SPSpeaker13:59

      Um, so I'm in my fourth year and currently in civil engineering.

   8. SPSpeaker14:03

      Great, and Adithi has brought you in to talk about the mech module. So jumping right in, um, I wanted to know: How different is it to design a rover for Mars than for Earth? What are the design challenges you've had-

   9. SPSpeaker14:16

      Hmm

   10. SPSpeaker14:17

       ... and how have you approached them?

   11. SPSpeaker14:19

       So, yeah, so, you know, there are quite a lot of differences

8. 14:22 – 21:18

   Designing a Robot for Mars: The difference and challenges

   1. SPSpeaker14:22

      when you design for Mars as compared to designing for Earth. And you need to take a lot of parameters into consideration, like environment, the soil condition, the area that you need to traverse, right? What kind of extreme conditions that the rover will have to face, because when you send a, you know, ma- a rover to Mars, it actually costs a lot, right? And plus, there's, there's an objective that the, uh, rover has been sent to, which it ha- it ha- it has to accomplish. Uh, comparatively here, you know the conditions pretty much well, right? And we operate more over on a Martian-type terrain, right? So the environment is just Earth-like, it's not what Mars actually has. So the material science, uh, you know, comes in a lot here. Uh, when you have one... When you have it on Mars, like, you have different materials, like titanium and all, that are used, which are quite expensive. But for, uh, you know, the Earth conditions, the materials cannot... We do not need to use such kind of materials. You can have it a much simplified version. And coming on the designing part, uh, a- as I said, again, the rover has an objective to be... You know, when it's on Mars, it has an objective to c- accomplish, and here we have a more over for a competition, uh, you know, to win competition, to participate in competition. So the design is quite different as compared to what you have on Mars, as compared to what you have on Earth.

   2. SPSpeaker15:40

      Right. Makes sense. And if you were to give me some highlights, um, on what you have achieved in the 10th iteration of Isaac versus when you were just starting out?

   3. SPSpeaker15:48

      Yeah. So we have, like, you know, evolved a lot, I would say, with innovations each and every year, you know, getting into new technologies. So, like, initially, we used to use pneumatic tires. Um, aluminum, uh, was the main, you know, co- component of our entire mechanical module, right? And, uh, even the electrical part was very simplified. Uh, sorry, very simple. We had- did not have any much innovations there. But, uh, over the years, we have, like, in- incorporated a lot of new things. Like la- since the last three years, we have incorporated, uh, 3D printing technology a lot. So now currently, our rover is around 30% of 3D printed, right? So it gives us a lot of room for experimentation, decreases the weight, uh, and also, you know, saves us some money. Ma- and over the 10... Over the last 10 years, like, I've been in the team since three years, so what I have noticed is each year we try to get some innovation in each and every model, module, I would say, and each model is im- a improvised version of what it was, uh, prior. So 3D printing, as I said, was one of the major things that we incorporated in mechanical this year, right? We also incorporated a steering, which was not there before. Uh, so that is a new innovation, which was, uh, incorporated again. So yeah, each year we try and do new things.

   4. SPSpeaker17:04

      I'd like to know more about both the 3D printing that you mentioned, like what kind of parts have you swapped out? Are they only static parts, or any moving 3D printing parts as well? And also about steering, what have you incorporated in steering?

   5. SPSpeaker17:16

      Correct. So, uh, in 3D printing, I would say, right, uh, mainly if you see our wheels are 3D printed, right? Our arm, uh, uh, the gearboxes are 3D printed, the fingers or the gripper is, uh, 3D printed. Again, this has been... You know, we achieved it over the last three, four years, so it took a lot of time to iterate and replace such parts. So last three years, we actually switched from an aluminum-based wheel, a foam-based wheel, to a fully 3D-printed, uh, wheel, right? Uh, slowly and steadily. And even the gearboxes, so we have included a cycloidal gearbox, right, which is completely 3D printed. And this 3D printing, we do it in-house, in our own workspace. So that is something that we have achieved over the last, what, three, four years. Yeah.

   6. SPSpeaker17:58

      I honestly did a double-take [chuckles] when you told me that the tires are 3D printed, 'cause they really don't look like it. And I don't know, the sort of 3D printing I've seen in labs usually is-

   7. SPSpeaker18:08

      Yeah

   8. SPSpeaker18:08

      ... PLA-based and not that durable. So I am genuinely very surprised that these tires have clearly sustained testing today.

   9. SPSpeaker18:16

      Yeah.

   10. SPSpeaker18:16

       I can see the mud on them.

   11. SPSpeaker18:17

       Yeah.

   12. SPSpeaker18:18

       But, like, is it-... like, why not use the same pneumatic tires that you had in the first iteration? Why swap to 3D-printed tires here? And can such tires be used for commercial vehicles on Earth, or is it specific to Mars again?

   13. SPSpeaker18:32

       So again, uh, I would say initially we started with pneumatic, as you said, right? But we faced a lot of issues there, like it's more over, you know, uh, there would be punctures because the Martian terrain is quite rocky, and what we have in competitions is also quite difficult to traverse. So, uh, because of that, you know, we faced a lot of issues and during competitions, while the missions are going on, you can't really actually change an entire wheel, right? So we slowly understood that we need to change to a better system, and I would say fo- uh, 3D printing was not the first idea. We iterated a lot on foam-based wheels and different kind of wheels, but finally, we landed on 3D printing because, uh, it was much more reliable, I would say, right? Uh, we haven't faced any issues with the pri- uh, wheels, particularly in missions, uh, since the last three years. So that was one of the main, I would say, you know, uh, reason why we switched into 3D printing wheels. And, uh, moreover, I would say, yeah, I, I feel, uh, like soon we would be mo- moving into 3D printing, uh, 3D-printed wheels, not particularly for, uh, you know, our commercial vehicles as such in the recent future, but into some smaller robotic, uh, vehicles, you know, or, uh, UAVs, as we said. Uh, a-along with that, as you asked about the Martian terrain, right? The, uh, 3D printing cannot be used on Martian terrain, if I say, because, uh... or I would say it can be used on Martian terrain, but not like on Mars- Martian environment, because temperatures are very lo- they go at extremes. We can't sustain that temperatures. 3D printing materials cannot sustain that, uh, uh, temperatures. And this is not metallic 3D printing. It is, uh, volume on 3D printing, right? So, and the regolith that is actually present on Mars is also very, uh, tough to traverse, as well as it's very sharp, right? As come- even though we try to replicate it on Earth, right, and different agencies try to, uh, you know, uh, test their rovers there, but it's not an exact replication of what we find on Mars. So that's why th- I don't feel the 3D printing, as of now, can be used in, on the Martian terrain, but yeah, innovation is... There's always room for innovation, so yeah.

   14. SPSpeaker20:38

       That was quite informative. Thank you. And, um, 3D printing, I suppose, would have massively saved costs versus going for other-

   15. SPSpeaker20:46

       Mm

   16. SPSpeaker20:46

       ... types of manufacturing processes.

   17. SPSpeaker20:49

       Yeah.

   18. SPSpeaker20:49

       So, and that brings me to the question: How much does it take to manufacture this kind of rover in-house?

   19. SPSpeaker20:56

       Yeah. So, uh, currently, each rover is around 3.5 to 4 lakhs, right? Slightly, uh, there's always, uh, some changes in there, but yeah, moreover, 4 lakhs, uh, we can consider. And 3D printing has actually significantly dropped our cost by around 20, 30%. So, you know, that cost is again put into innovation, and it's, uh, like a flywheel. So as I said, steering we incorporated,

9. 21:18 – 25:30

   Money Matters: Rover costs and saving costs with 3D printing

   1. SPSpeaker21:18

      we have cycloidal gearboxes. All of those, there was room for that, uh, because we saved costs through 3D printing, so.

   2. SPSpeaker21:27

      In the competitions, one of the crucial aspects is navigating the terrain, as we've been talking about. Has there been any particularly challenging terrain that you faced during the competition that's maybe sent the rover almost for a tailspin or left you very scared?

   3. SPSpeaker21:44

      So yeah, one, in one of the missions, uh, in Turkey, which we had a competition, was that we had an astrobiome mission, and, uh, uh, I was the runner, so I was seeing the rover running and traverse, navig-navigating the entire terrain. What happened was suddenly, one of our drive motors bursted. There were sparks and spo- uh, smoke that came out. So the organizers, for safety measures, you know, just, uh, stopped, uh, abandoned the mission. And, uh, I was the one who had to assess the entire situation. So I went there, and I was assessing why the drive, you know, suddenly just, uh, caused issues. And to my surprise, after assessing for 15, 20 minutes, every important thing, I found out that there was one bolt which was very loose, which was- which just came out and was causing resistance to the rotation of the wheel. And, uh, because of that one, and we lost one entire mission, which was very crucial for us, which could have, you know, improved our position by three to four ranks. So yeah, that was a very, I would say, you know, crucial moment because what the team learned was how small things actually matter. One screw not tight and, or one wiring going wrong could actually, you know, cost you a lot of things because you are, uh, you are indirectly show- uh, you know, representing your country in international competitions. So these are some points that, you know, actually define how, or actually tell us how much better we need to do the next time.

   4. SPSpeaker23:06

      And like you mentioned, how these small things, very small things, can have such a big impact. It is remarkable that so many subsystems working together, so many small things that could go wrong, small, big, all of that, you still have such an amazing functioning rover.

   5. SPSpeaker23:21

      Yeah, so we are... It's always, you know, we need to actually work a lot in team. A lot of collaboration goes on intermodular, because if electrical is there, so the mechanical has to work well. If mechanical is working, astrobiome has to work well. So all three subsystems when worked together, and we actually have, you know, regular intermodular meets where people discuss about different modules, what's going on, and how each module has to actually work. Uh, because let's take just mechanical, right? You need to change a lot of designs, keeping in mind about electrical wiring, astrobiological module. So, uh, we have those meetings specifically to understand that, okay, uh, for the astrobiological module... astrobiology module requires this, has its requirements, how can mechanical module or electrical module accommodate that? And that's how, you know, we are functioning as a team o-overall.

   6. SPSpeaker24:10

      Right. Not only intra-team, but inter-team communication-

   7. SPSpeaker24:14

      Correct. Yeah

   8. SPSpeaker24:14

      ... to come up with a cohesive rover-

   9. SPSpeaker24:16

      Yeah

   10. SPSpeaker24:16

       ... is very important. And that brings us to the next few modules. I want to know a bit more about the electrical, software, and astrobio modules.

   11. SPSpeaker24:24

       Yeah. So we have, uh, one of our team lead, Soham, who, you know, spec- who, who's the electrical and software lead of Team Anveshak. He handles the entire electrical and software for the rover, as well as for the drone module.

   12. SPSpeaker24:36

       ... Hi, Soham. Um, which branch and year are you from?

   13. SPSpeaker24:39

       So I'm in the mechanical engineering branch, and I'm in my third year right now.

   14. SPSpeaker24:43

       And you work on the electronics and software module?

   15. SPSpeaker24:46

       Yeah.

   16. SPSpeaker24:47

       Great. So, um, Ayush actually mentioned that you're also looking into the drone and rover collaboration. From what I understood, in the competition, you already have two modes of operation. You have the operator-led control and the autonomous drive. What I want to understand more is where does the drone fit in, and how does it help?

   17. SPSpeaker25:07

       So for the competitions that happen in India, the drone and rover are in a separate competition. So right now, m- a lot of Indian teams have not reached the stage where the drone and rover can work together autonomously. So for that, the drone works on a completely separate competition known as ISDC, the International Space Drone Competition, and where the drone has to go at various heights, get some sensor readings, and the drone is controlled

10. 25:30 – 27:31

    The Anveshak Drone: Masterclass on drones and how they work

    1. SPSpeaker25:30

       by an operator who, again, can't see the drone to simulate Mars. But in international competitions, they are expected to work together, where, first of all, the rover will go very far from where it is being controlled, and in those cases, the drone can act as an antenna in the middle for the base station to communicate with the rover. It'll be in between the base station and the rover, and it can k- transfer commands for the rover from the base station. One more thing it has- it can do in competitions is that the rover has to go and find stuff that will be lying around. So the rover is comparatively much slower than the drone, so the drone can go around and circle the entire arena and find out where the objects are and what is the easiest path towards them, which the operators can then take from the drone, and the rover can go and find those obstacles. So that is how in competitions they work together. And on the actual Mars rover and drone that are operating on Mars, it is mainly used so that you don't want to lose so much money that you've put into a rover. There's almost $50 billion worth of material up there, and you don't want to accidentally drive it into a ditch, and there's only so much you can see on a camera from the rover. So the drone will go, like, way further off than the rover and tell the people on Earth that what is there up ahead of the rover and how it should plan its next path.

    2. SPSpeaker26:41

       Right, so it's a collaboration between the operators, the drone, as well as the rover.

    3. SPSpeaker26:45

       Yeah.

    4. SPSpeaker26:45

       So the drone is acting like, you know, going around in circles, the eye, basically, the whole map, and mapping out the space so that the operators can make better choices and the rover can explore.

    5. SPSpeaker26:55

       Yeah, pretty much.

    6. SPSpeaker26:56

       Great. And autonomous driving, like on Earth itself, is a challenge.

    7. SPSpeaker27:01

       Yeah.

    8. SPSpeaker27:01

       It's something we're still figuring out. So on Mars, how are you guys figuring it out, and how much of it really is autonomous if you also have the operators on board?

    9. SPSpeaker27:09

       Okay, so the first, one of the biggest challenges to autonomy on Earth is that it's a dynamic environment. Everything around us is moving, which is not the same on Mars. So it actually makes autonomy on Mars much easier compared to autonomy on Earth. And also, again, it's not fully autonomous on Mars as well, it is semi-autonomous. There's an operator who will approve everything that the rover does before

11. 27:31 – 32:17

    Autonomous Driving: How do you remotely drive a vehicle on a different planet?

    1. SPSpeaker27:31

       it actually does the autonomous part. And yeah, so on Mars, what the rover will do is basically after the operator gives it a command... Let's say the operator says, "You need to get here." Then the rover will use the sensors to plan a path to get there, along with what the drone is telling it, and after it finds the best path, it will then send the path back to the operator. The operator then will verify this path, and the rover will then start moving along this path. It will go very, very slowly, and then in case there is some obstacle which was not in the map or not captured by any of the sensors of the rover, the rover will dynamically plan a path around that obstacle in real time. That is how autonomy works on Mars. On Earth, it's a completely different thing because everything is moving, and you need to think of the safety of the people around you as well. On Mars, that's really not a concern.

    2. SPSpeaker28:19

       Right. On Mars, we're concerned about our rover. [chuckles]

    3. SPSpeaker28:21

       Yeah, that's it. As long as the rover is safe, it's all fine.

    4. SPSpeaker28:24

       Yeah. So to power this autonomy, like, of course, you need a lot of sensors, I believe-

    5. SPSpeaker28:29

       Yeah

    6. SPSpeaker28:29

       ... and as you mentioned, real-time processing, dynamic path planning. How have you guys architectured this?

    7. SPSpeaker28:35

       Okay, so for, for only the autonomy part, we have a lot of sensors. We have a stereo camera called Z2I, and we also have a LiDAR, and as well, we have, we have a GPS sensor and an IMU as well. So all of these, combined with the electric architecture, are powered by something called an SBC, a single-board computer made by NVIDIA. It's called an NVIDIA Jetson Orin Nano. Okay, so this thing is around the size of my palm, and it's probably more powerful than the laptop that you're watching this on right now. It is a CPU plus GPU combined, which can process all these sensors together and run a lot of machine learning and AI models to give us perception and depth of everything around the rover. That is the broad autonomy stack.

    8. SPSpeaker29:20

       Okay, great. Yeah, I have heard about the Jetson Nano being really, really powerful and NVIDIA releasing it and helping labs and research teams-

    9. SPSpeaker29:27

       Yeah

    10. SPSpeaker29:27

        ... across the world basically, um, have localized, you know, GPU compute, rather than having, like, the whole bulky, um, system or connecting to a server, for that matter. So this was, like, all the higher level and all the sensors and subsystems communicating with each other. What about the lower-level electrical architecture that you have?

    11. SPSpeaker29:49

        Okay, so for our lower-level architecture, the rover is, first of all, powered by two 24-volt batteries, which are around 24,000 mAh each. So these batteries, one battery is used to power the onboard computer, which is the Jetson Orin, and the arm. The other battery is used to power the drive. So the arm is like the robotic manipulator we have, and the drive is basically the four wheels on the rover. Each wheel can... has a motor which controls the rotation of the rover, uh, of the wheel, and also another motor which controls the steering of the wheel. That is, each wheel can rotate in place itself. All right, so each PCB, which is basically a printed circuit board... You will have seen circuits, right? So printed circuit board is basically, these circuits are etched onto a plate. These etches are called traces, which are basically wires, but very thin, so that everything is compact and remains in place when the rover is on very rough terrain. So each PCB has a microcontroller, which is like a-... not a computer as such, but it controls the sensor input that is coming from the PCB onto it, and it sends this information to the Jetson Oren. So that is how communication works. It goes from the Oren to the microcontroller. From the microcontroller through the PCB, it'll go to motor drivers. All right? So motors need a lot of current to run. You can't just directly connect them to a microcontroller, which takes very little current. These motor drivers will handle the primary current exchange and control the motor. So from the microcontroller to the PCB, to the motor driver, to the motors, is how the lower level, uh, architecture works. It's the same for the drive and the arm. We have a separate PCB, which runs the drive, and another PCB running the arm. So the motor drivers are custom-made. They are made by the team itself, because outside, when we buy motor drivers, we face a lot of issues. Like, there might be reverse polarity at times, where you, by mistake, say, you plug in the motor driver in the wrong configuration. Now, where plus was, there's minus now, and the motor driver might just fry. You do not want that happening in competition, right? So we engineered the motor drivers ourselves to prevent all these issues from coming up.

    12. SPSpeaker31:46

        You guys have been doing a lot of stuff in-house, I guess. The 3D printing is in-house.

    13. SPSpeaker31:49

        Yeah.

    14. SPSpeaker31:50

        Like you mentioned, making the motor driver is in-house.

    15. SPSpeaker31:52

        Yeah.

    16. SPSpeaker31:53

        Anything else which you guys have been developing on your own?

    17. SPSpeaker31:55

        So yeah, we do a lot of things in-house. The motor drivers, like I mentioned, are in-house. The PCBs are completely designed by us. We just send them for manufacturing outside. A lot of the b- manufacturing that the mechanical team does is done by our warehouse itself, so we don't go anywhere outside for that as well. The 3D printing, as mentioned before, is done by us. Then for the software stack, around 50%,

12. 32:17 – 33:50

    In-House Development: What parts has the team built on its own?

    1. SPSpeaker32:17

       we use packages built by other people. A lot of ... We like to prep, uh, s- what is it called? We like to promote as much open source development as possible, while the remaining 50% is written completely by us. Then the astrobiome team, they use a spectrometer, which is, again, completely made by us.

    2. SPSpeaker32:34

       That's very interesting. I actually want to find out more about this astrobiome team, and I guess that's the final piece in the picture, right?

    3. SPSpeaker32:40

       Yeah.

    4. SPSpeaker32:40

       You have the mechanical team, which has the whole chassis. Then you have the electrical and software subsystem being the brain of the robot controlling it, and finally, the main goal is to get the samples from Mars.

    5. SPSpeaker32:51

       Yeah. So the astrobiome team works on the last part that you said, that is getting the sample from Mars, and they are analyzed on the rover itself. So it'll collect the soil, and the rover will act as a mobile science laboratory to analyze these samples. For astrobiology, it's better if Abhishek comes and explains that part now.

    6. SPSpeaker33:08

       Hi, Abhishek. Uh, which year and branch are you from?

    7. SPSpeaker33:11

       So I am from the Metallurgy and Materials Engineering branch, and I'm from the third semester.

    8. SPSpeaker33:16

       And you're leading the astro-

    9. SPSpeaker33:18

       No, I'm, I'm a member of this, uh, module. I joined, uh, like, last year.

    10. SPSpeaker33:24

        Okay, so can you tell us a bit more about the astrobiome module? What do you do?

    11. SPSpeaker33:28

        So astrobiome module, our main mission is to collect soil sample and then store it, and using that soil sample, we are trying to analyze if that soil sample has certain things that s- that suggest that life, whether it exists now or could have existed in the past. So that is our, uh, mission. So how we do this is basically we use a drill

13. 33:50 – 37:43

    Anveshak Astrobiology Module: What is the mission and what does the module do?

    1. SPSpeaker33:50

       mechanism to collect the soil, and then we have a collection box in which we store it in. Once we do that, we have, uh, built a 3D-printed in-house spectrometer. So using that spectrometer, what we do is that we search for signs of life, like protein or carbohydrate, in the soil. So this is what we basically do.

    2. SPSpeaker34:08

       And yeah, you mentioned the in-house spectrometer again, so why not buy an off-the-shelf product? I imagine-

    3. SPSpeaker34:14

       Mm

    4. SPSpeaker34:14

       ... building a spectrometer would be very challenging in itself.

    5. SPSpeaker34:17

       Yeah. So the thing is, an off-the-shelf spectrometer is usually very costly, like a lakh or so. But this spectrometer that we built, because we are using parts that are, like, cheap, and we are 3D printing it as well, we are, we are- we can reduce the cost so that it's, like, some 30,000 or something. So it's very cost-efficient, and it's also like a engineering challenge to us to do it, so it is a very interesting thing.

    6. SPSpeaker34:41

       And all of this post-processing that happens, is it done on the rover itself, or do you take it back to a lab and some equipment is designated?

    7. SPSpeaker34:48

       No. Uh, so the, all this, all the spectrometric tasks all happen in the rover itself. The data alone is ... Then we get, once the spectrum alone we get, which we analyze it, uh, uh, and elaborate.

    8. SPSpeaker35:00

       So when it comes to drilling, how do you decide which spot to drill? Is that known beforehand, or does the rover intelligently decide that this looks like a good spot to drill?

    9. SPSpeaker35:10

       So before competition itself, we will be shown the area, the competition site. So we will choose the spot based on, uh, feasibility, we will choose a spot based on, uh, where, what type of soil is there, and whether or not we expect certain biosignatures. So how we do, we also have a ML model that classifies, uh, and identifies the rocks. So we, using all this data, we will be able to figure out what spot to drill in, and then we will be able to, uh, find any biosignature, if it exists or not.

    10. SPSpeaker35:41

        Okay, so currently, when you're testing, um, what kind of life signatures do they give to you in the competitions to look at?

    11. SPSpeaker35:48

        So basically, the main things that we are trying to detect are proteins, sugars, fats, and et cetera, and so on. So basically, the soil, it depends on where the competition happens. This time it's happening in Udupi, which is majorly a desert region. So based on that, the type of soil varies, and the type of things that we are trying to identify as well varies. So beforehand itself, we will have a basic idea of what we are trying to find based on where the competition is happening, and using that, we will try to ... We will choose appropriate, uh, appropriate, uh, chemical tests and so on to find.

    12. SPSpeaker36:21

        Do you need a lot of, like, apart from the ML knowledge that you already said, even like chemical, chemistry, and biological knowledge to be able to function as a module?

    13. SPSpeaker36:31

        Yes. So basically, we have a lot of chemical tests as well to detect, uh, molecules. Uh, for example, there is a chemical called Benedict's reagent, which we use to fi- figure out if the soil contains-... sugars in it. So, uh, similarly, we have several other tests. So to do well in this, we have to have a good knowledge of both electrical stuff, because we have to make the sensors work, good mechanical stuff for the drilling system, good software stuff for the, uh, ML model to work, and good biology and chemical knowledge as well. So it's a very interdisciplinary module that we have. So using all this knowledge only we'll be able to proceed in this mission.

    14. SPSpeaker37:09

        Seeing Adithi, a fellow woman in STEM, I was very inspired. Not only is she in a technically demanding competition team, she's leading it. So Adithi, I wanted to know what drew you to STEM in the first place?

    15. ADAdithi37:22

        So during my high school preparation itself, I realized that I was more inclined towards the physics and ma- maths in the subjects. And apart from that, also, both my father and my brother are, you know, working in mechanical engineering field. So I was naturally drawn towards the STEM field, and I was

14. 37:43 – 43:09

    Women in STEM: Adithi’s advice to girls in STEM and future plans for Anveshak

    1. ADAdithi37:44

       very much resonating with the, uh, uh, with the field of engineering. So that is what I chose to do and take up my career path forward in that.

    2. SPSpeaker37:53

       And was there any resistance for you taking up a mechanical-led, um, study path? 'Cause generally there's the stereotype as you...

    3. ADAdithi38:02

       Yeah. So there was a bit of, you know, hesitation before, you know, selecting mechanical engineering in particular, because, uh, a more of a generic trend that people would usually assume girls to take up would probably be engi- uh, like electrical engineering, even if people, like girls, chose STEM as a field. But, um, like I said, that I had, like, full support from my family, and they understood that I had, like, a keen interest in the domain of mechanical engineering. So I was fortunate enough to not really be subjected to the stereotypical thinking as such, but there... I would say there was still a bit of hesitation that I am happy that I was able to pass through.

    4. SPSpeaker38:48

       And now, almost four years done with college, three years in the team, is there any advice you would give to your younger self or to a fellow girl entering STEM?

    5. ADAdithi38:59

       Um, so an advice that I would like to give is to focus on what you like more than what other people would probably think about you. Because at the end of the day, it's about what makes you feel happy. So for me, it was choosing a career path in the field of engineering, because that was something that looked very exciting, even though that probably was, uh, something that was very, you know, probably chaotic down the line. But that is something that I chose to do, and I would encourage my younger self to choose in a similar way, and I would also encourage the other, uh, fellow women in STEM, to keep pursuing their dreams and not to restrict yourself from d- achieving something that you truly want to.

    6. SPSpeaker39:47

       And coming to Anveshak, you've had the whole journey, right? From a second year to a third year, being the engineer on the team, to finally now team lead. As a personal preference, what do you enjoy more, managing and leading the team or being the engineer on the team?

    7. ADAdithi40:01

       Um, to give a diplomatic answer, I would say I like doing both. But to be very realistic, it is the engineering part of it, because the, uh, thrill of having to work on something very new and having it to be integrated onto something as robust as a rover and actually getting to see it work, I think that is w- uh, more of where my interests lie in. So I would say that I liked the engineering part of it more. But then again, as we, uh, improve in our career path, there is at a time where you have to guide your juniors, and so that is also something that I'm growing to like a lot more now down the line.

    8. SPSpeaker40:44

       That's great, and I'm sure you're inspiring a lot more juniors along as well. One more thing, like you mentioned, right now, you guys are doing testing every day, and on average, you spend around four hours a day over here, which is very technically demanding and time-consuming, plus we have our academics in IITM. So a general trend is that there are many people who join these competition teams, but maybe in a year or two, they leave. You have stayed in the team, and now you've moved on to lead the team. So what has kept you in Anveshak?

    9. ADAdithi41:15

       So that's very true in saying that there were a lot of hardships along the way, just in the first two years into the team. But the main reason that made me, I guess, continue into a third year was the... just as cliché as it sounds, the generic love for the team, and that one vision for the team to, like, guide it into a particular direction, seeing the potential that we have within the juniors, and well, rightly channeling it and utilizing all the opportunities that we have, and being able to explore into multiple space technologies that we do right now. That is- That just felt so enriching in itself, that made me decide to stay another year and try my best into incorporating certain changes that would make us like, a better team than we were before.

    10. SPSpeaker42:04

        Okay. Thank you, Adithi. You've left me feeling quite optimistic about the future of women in STEM. And now coming to Team Anveshak, what is the big picture or the future vision towards the team? What are you building towards at the end of the day?

    11. ADAdithi42:19

        So the big picture that we have in mind for Team Anveshak is to basically develop further in space technology, and also to eventually be one of the organizations that would send one of our rovers on Mars and explore the unexplored.

    12. SPSpeaker42:34

        Thank you, Adithi, for your and your team's time. In today's episode, we saw how Team Anveshak has combined various modules, electrical, mechanical, software, chemistry, and biology, all into one power-packed robot running in tandem with a drone to automate and accelerate space tech exploration. I believe it is research from labs and competition teams such as this, which is providing India the boost to move ahead in the space tech race. If you like content like this, do like, share, and subscribe, and see you in the next episode.

Episode duration: 43:14