Developing Algorithms to Promote Independent Navigation and Mapping in Unknown Environments
The Intel® Student Ambassador Program was created to work collaboratively with students at innovative schools and universities doing great work in the machine learning (ML) and artificial intelligence (AI) space. Khen Elimelech has focused his research on autonomous decision-making and reducing the computational demand of complex AI problems.
Tell us about your background.
I was born and raised Tel-Aviv, Israel (a.k.a. the Silicon Wadi). I completed a B.Sc. in applied mathematics while in high-school. I then joined the technological arm of the Israeli navy for about 5 years, and served as an officer and team-leader, while pursuing an M.Sc. Now, at the age of 24, I am a full time Ph.D. student in the Program of Robotics and Autonomous Systems in the Technion – Israel Institute of Technology. I feel that this program reflects best my interdisciplinary interests in science and technology.
What got you started in technology?
I have been immersed with technology and computers for as long as I can remember. I was lucky to grow up while the world was going through the Internet revolution. When PCs at home were still a new concept, I was teaching myself web programming and building websites. I had an online portal for teens, and it enjoyed a decent following, way before social media. It might sound basic in today's terms, but for a 10-year-old back at the beginning of the century, it was a big deal. Later on, I progressed to more serious coding of programs, and digital graphic design.
What projects are you working on now?
Currently I am working on completing my Ph.D. thesis. My focus is on developing fundamental algorithms to allow efficient autonomous decision-making in real-time. These ideas can be applied to various tasks, such as autonomous navigation and mapping, manipulation of robotic arms, real-time classification and many more. I now work on implementing these algorithms on real robots, which perform autonomous navigation and mapping in unknown environments. I expect to see a significant improvement in efficiency, after already achieving promising results in simulation.
Tell us about a technology challenge you’ve had to overcome in a project.
Usually I start by examining mathematical theory and algorithms before reaching for the computer to code and verify how they perform. Many times, I discover that ideas which are true in theory cannot be applied in practice, because of technological limits – the information becomes too large to handle, limited precision, or lengthy processing time. The problem is not only about solving an equation but solving it in a way that is practical within the technological constraints. This is actually the underlying idea of my research – reducing the computational demand of complex AI problems.
What events have you attended as an Ambassador?
I was fortunate to be invited to the Student Ambassador Summit earlier this year, which took place in conjunction with the Intel® AI DevCon (Intel® AIDC) in San Francisco, California. It was a wonderful experience! I got to meet fellow Ambassadors from all around the world. It was interesting to see how diverse the group was and hear about each person's background and projects. We each showcased a poster about our work, and I was happy to get a warm response for mine. Alongside forming a community of like-minded friends, I hope we will also be able to collaborate in the future, in one way or another. Intel AIDC was different than the academic conferences I previously attended – definitely a step outside of the academic bubble. It included many hands-on sessions and provided good exposure to real applications and interests of the industry.
How can Intel help students like you succeed?
Intel can help students by giving access to state-of-the-art hardware and products, which might not be easily available for students. Also, a collaboration with Intel can expose students to the work environment in a major technological company and help them understand the industry needs. Through events and conferences, Intel can also help students in creating new connections with relevant people.
What impact on the world do you see AI having? And do you see yourself as part of it?
These days we cannot imagine our life without the internet and smartphones, and I think AI will have a similar destiny. Almost every aspect of our life can be improved by AI. Various rigorous tasks are destined to be replaced by robotic agents for the sake of efficiency. The Internet of Things (IoT) trend is already aiming to bring autonomy to everyday devices. There is also a huge push of using AI in medicine and agriculture, and of course the topic of autonomous driving. I think that in the not-so-far future we will view hopping into an autonomous taxi, or performing a medical procedure by a robotic arm, as a thing of the ordinary. Being in the mix of things in the right time, I definitely see myself a part of this revolution. Especially with the focus of my research, which I hope can help extending the reach of AI.
Outside of technology, what type of hobbies do you enjoy?
I like staying active by practicing gymnastics, yoga, diving, and cycling outdoors - always accompanied with good music on my headphones. I am also an avid traveler, having visited over 30 countries in 5 continents and counting.
What are you looking forward to doing with Intel?
As part of this collaboration with Intel I hope to take my ideas outside of the university walls and examine if they can be applied on actual products. Also, I definitely look forward to participating in more Intel events, and hopefully host a local AI event of my own in the near future.
Join the Intel® Student Ambassador Program for AI
Upon acceptance into the Intel® Student Ambassador Program for AI, Graduate and PhD students from top universities worldwide can access newly-optimized frameworks and technologies, hands-on training, and technical resources provided by the Intel® AI Academy. To be considered for the program, apply today.
Product and Performance Information
Performance varies by use, configuration and other factors. Learn more at www.Intel.com/PerformanceIndex.