At TUMCREATE, as we prepare to conclude our research in Phase Two – Towards the Ultimate Public Transport System, we would like to highlight our researchers who continue to make significant contributions to the programme. Through our interview series, People behind the Science, we talk to our researchers to discover more about them and their stories.
We talk to Dr Alok Prakash, Principal Investigator of the Sensing and Management for Agile Transport (SMAT) team.
Alok grew up in a small city in India, called Bokaro Steel City, where true to its name, is famous for its steel manufacturing plant – a fun fact that Alok proudly shared with us. Growing up, Alok was such a fan of Sci-Fi movies that he never missed a single one of them, especially the ones with the word ‘Star’ in it. Admitting to being a ‘tech nerd’, he is always ready to lecture anyone on the latest and upcoming gadgets when given the chance. He also confessed to being a frequent Carousell user in order to keep up with his hobby of changing his phone every month. That is definitely a hobby that only a passionate ‘tech nerd’ will have!
Q: Tell us about yourself and what you do…
I am a Senior Research Fellow (SRF) in Nanyang Technological University (NTU) and the Principal Investigator (PI) for the SMAT team within TUMCREATE since February 2018. Similar to other PIs in TUMCREATE, my work encompasses a diverse range of tasks such as research management, administrative tasks for the team, liaising with national and international academic as well as industry partners and planning, coordinating and overseeing the several ongoing field trials within the NTU campus and beyond. As an SRF in NTU, I also guide PhD students in their research. Additionally, I also lead the hardware design and prototyping team within SMAT that is responsible for maximizing the performance of low cost, low power edge computing platforms used in our sensors through sophisticated mapping and optimization techniques.
Q: What made you come to Singapore, and how has it been like living here?
After finishing my undergraduate degree at the Indian Institute of Technology (IIT), Dhanbad, India, I was looking for an opportunity to pursue higher studies overseas. I had applied to several Universities in the US and received an acceptance offer from the University of Michigan at Ann Arbor. I was over the moon to have received the acceptance letter so imagine my disappointment when I realized that I would have to secure a massive loan even to apply for a US VISA. This was when one of my friends, who had been working at NTU as an intern, advised me to try for a Masters/PhD position in NTU. Since the application deadline had already passed, I was offered to come to NTU as an intern first with a possibility of starting my PhD from the subsequent semester. I accepted the offer and I still remember the exact date that I landed in Singapore, 2nd August 2007.
Thirteen years later, in retrospect, it was one of the best decisions of my life and it has been the most enriching experience. During the course of my PhD, I had an opportunity to visit the UK for three months as an intern and I missed Singapore so much! Ever since then, I have made it a point to not stay away from this country for more than two weeks at a time or I'll be homesick.
Q: Who or what inspired you to be in your field of research?
I have always been enticed by gadgets and electronic items used in our daily lives. Growing up in India, though we did not have access to many, I was always trying to collect and fix them when something did not work as expected; sometimes successfully, at other times not so much. When the time came for me to select my major for undergraduate studies, I decided to pursue Electronics Engineering which was in line with my passion. My PhD and post doc research topics continued in this direction with a sharper focus on low power embedded computing platforms that are ubiquitous in devices around us.
Q: What are some of the challenges for you in your field of research?
With the advent of smartphones, the capabilities of embedded systems have grown many fold. There has been a tremendous growth in their computing power in the last five years alone. However, while the hardware capability has grown almost exponentially, the growth in the accompanying software needed to make optimal use of all this computing power has not kept pace. As such, we have a situation known as ‘design-productivity gap’ that necessitates more work on topics such as automated partitioning, mapping and optimizing algorithms to make the best use of the underlying hardware.
At the same time, the unprecedented rise in Artificial Intelligence (AI) and Machine Learning (ML) applications, which were traditionally deployed on much more powerful platforms, but currently geared towards embedded systems such as smartphones and IoTs, now demands more sophisticated mapping and optimization strategies to ensure high performance without violating their strict power and thermal constraints.
Q: Tell us about your work and journey in TUMCREATE…
I joined the SMAT team as a research fellow at the start of TUMCREATE Phase 2 program in 2016. My primary focus at the time was to develop automated techniques for mapping complex applications, such as the novel computer vision and AI applications that were under development in SMAT, on low cost and low power embedded computing platforms while ensuring high performance. I was promoted to become the PI of SMAT in 2018, just a few months before a critical review of the program, which significantly broadened my responsibilities almost overnight. It was an excellent opportunity for me to stretch myself beyond research and learn several other useful skills like leadership and management.
Since then, SMAT has developed several state-of-the-art AIoT sensors that can be easily scaled for city-wide deployment. Our sensors are capable of addressing key issues faced by current transport systems such as the lack of priority for public buses at traffic lights, private vehicles entering dedicated bus lanes and illegal parked or stopped vehicles that impede the smooth flow of traffic. We have also developed novel rapid scheduling techniques to enable demand-aware scheduling for public transport buses that react to real-time demands at bus stops. Most of the AIoT sensors have gone through extensive field trials conducted in the NTU campus and I am happy to inform that they are now ready for large scale deployment.
As the PI of SMAT, the journey has been extremely satisfying with SMAT’s contributions being widely recognized by both academic and industry partners for its potential in solving several transport related challenges as well as lucrative opportunities for commercialization.
Q: What are some of your plans in the future?
I am not going to leave Singapore ever, if that is what you mean! This is one thing that I am certain of. Regarding my career, I plan to join academia at some point in the future where I can have the freedom to define my own research path while also training the next generation of researchers in this exciting field.
(As a self-proclaimed tech nerd and a scientist who works with cameras and sensors, how will Alok feel about a camera that is able to store emotions as well?)
Q: If there was a camera that could capture a past memory of yours, including the emotions felt, so you can relive it again, which memory would you choose?
Honestly, I am a person who does not like to dwell on the past but if I have to choose, it would be some of my fondest memories. When I cracked the IIT-JEE exam in 2002, when I finally came back to Singapore after almost a year of being stuck in a hospital due to some health complications and the trip to Japan last year filled with numerous rides on amazing trains. It would be nice to relive and feel all those emotions again, especially the last one on my Japan trip since it is basically impossible to travel this year with the ongoing pandemic.
It's great to hear from the man behind our state-of-the-art AIoT sensors! We can’t wait to hear more successes from the team and see the sensors deployed city-wide.