An Aerospace Geek And Roboticist
Intelligence and Autonomy Engineer @ XOsight B.V
Enschede Area, Netherlands
He is a former Research Assistant at the Unmanned Aerial Vehicles Laboratory, Department of Aerospace Engineering at the Indian Institute of Science (IISc) with experience in mathematical modeling of complex flight vehicle dynamics and expertise in robotic perception-based navigation.
Navaneeth is certified in Artificial Intelligence and its applications by the IISc.
What inspired you to pursue a career in Robotics?
I got into the sphere of robotics completely by accident. I graduated with a bachelor’s degree in Aerospace engineering in 2016.
Being a niche industry, there weren’t many companies actively hiring for aerospace roles and those that did require a vast skillset that a mere bachelor's program didn’t offer.
I looked for ways to improve my skill set that would make me employable and I found that opportunity at a UAV research lab. Projects there required a deep understanding of the aerospace concepts as well as knowledge to practically implement these solutions.
I got into robotics as a way to bridge the gap between concepts and their implementation.
As I learned more, I realized I enjoyed robotics as much as I enjoyed aerospace and voila! I am here pursuing my master's degree in robotics.
What exactly an Aerospace Engineers do?
Aerospace engineers are basically the embodiment of “jack-of-all-trades”.
For example, the knowledge of materials engineering is required to understand the stresses the vehicles go through in a mission, the knowledge of electronic systems and computers go into designing a guidance/navigation system, the knowledge of fluid mechanics go into designing the aerodynamic aspect of the craft and so on.
We generally try establishing a base understanding of all the fields but choose a field we would like to specialize in and gain a deeper understanding of the chosen field.
So the saying would go like “Jack of all trades. Master of one”
Could airplane pilots be replaced by Robots in the future?
Definitely. Drones are proof that such a vision was already in the works from back in the 1950s.
Nowadays, every commercial aircraft has an autopilot that can land itself with centimeter-level precision without pilots’ intervention.
Regulations and culpability issues prevent aircraft manufacturers from going all out replacing pilots with such sophisticated systems.
Mention any interesting projects related to UAV?
My favorite drone project was the one that seems to be taken right out of science fiction. It was a project to control a drone with just your thoughts.
I was the test pilot for that project and I was made to wear an EEG headset that recorded electric fields from different parts of my scalp which together let the drone know which direction I wanted it to go.
It was a surreal experience when the drone took off the first time with just a thought of “TAKE OFF”.
Have you built from scratch an Unmanned Aerial Vehicle?
Oh yes! It was just a simple aircraft made using foam and hot-glue.
I wish I told you I designed it myself with aerodynamic calculations and stability analysis, but I just took the template off the internet.
That was when I learned building a good machine is not enough. You need to learn to fly it too.
However, last year I (with a group of friends) built a drone literally from scratch.
Calculations, designing, simulations, molding, and everything. It took about 8 months. Usually, it would take lesser time but as I was also busy with my master’s classes.
Anyway, the end result was something I am always proud of.
Explain how does UAV autonomous system works?
A completely autonomous UAV has three different layers of control.
The lowest layer controls the basic control surfaces or throttle that ensures the UAV is flying in a stable configuration.
This layer, however, doesn’t care about where the drone is which would be a problem in case there is the wind that might push around the drone.
Only this layer of control is required if you are piloting the drone by yourself.
The second level of control ensures the drone is flying at the location it is commanded to. This level makes the drone resistant to wind disturbances.
The third layer of autonomy would bring in the navigation algorithm which tells the drone what to do and how to go about achieving that objective.
This type of three-layer autonomy will let the drone operator press a button and let the drone do everything and come back.
Which Aerospace manufacturers company would you love to work for?
My dream since graduating was to work for Airbus.
The company is known for its innovative approach to improving the efficiency of its aircraft and ever upgrading its technology suite keeping pace with the passenger demands and hospitality.
As each aircraft has a life span of 20 years, they go on a technology overhaul every 5 years which is a daunting task and the innovation done by airbus to future-proof their products blows my mind every time.
Explain the difference between a Drone, UAV, and a Quadcopter?
The drone is a very general word. Though the word elicits the image of a flying DJI phantom, the technical translation of the word is Remotely Piloted Vehicles (RPVs). These can be anything, a flying vehicle, an RC car, or an unmanned submarine.
Any vehicle regardless of the mode of translation is controlled by an operator not physically inside the vehicle is a Drone.
UAVs are a subclass of drones. Any drone capable of flying is called a UAV. This encompasses fixed-wing, rotary-wing and lighter-than-air categories of aerial drones. A quadcopter is a subclass of UAVs. It falls under the category of rotary-wing UAVs.
A quadcopter is a UAV that propels and controls itself with the help of 4 motor powered propellers.
Do you have any Robotics project that wants to share your experience?
Currently, I am immersed in my Master Graduation Project.
The aim of the project is to have an eye in the sky looking for speeding vehicles and crashes on freeways and highways.
The end goal will see a Vertical Take-Off and Land (VTOL) UAV carrying a small but powerful computer running a neural network that observes the traffic below and reports all speeding violations and crashes it detects.
This is a very ambitious project for me due to the number of ideas currently in the field. I have to find out what works not only on paper but also works great in practice 200ft in the air. This is a perfect blend of theory and practice.
I also have amazing support from my supervisor and other Ph.D. students in the UAV group at my University.