Chapter 1: Maral's Story on Self-Driving Cars and Their Effect on People’s Life
In my last blog post, I promised starting a new series called Iranian Women in Science & Engineering, aiming to find Iranian women researchers scattered in the world and hear out their story and passion for advancing in Science and Engineering. The first person who kindly accepted my invitation to participate in this series was Maral Khosroshahi. Maral is a Machine Learning and Imaging Scientist at Konica Minolta Laboratory, in San Mateo, California. She received her B.Sc. in Electrical Engineering from Sharif University of Technology in 2010. She received her M.Sc. and PhD in Electrical Engineering form Illinois Institute of Technology in 2012 and 2017. Her research focus is on Computer Vision, Machine Learning and Deep Learning. Maral is also the Founder and Chair of Iranian Women in Computing, an Anita Borg Institute community which aims at building a strong network of Iranian women technologists around the world to improve their impact and visibility on the Tech community. Below is her story on self-driving cars and their effects on people's life.
Every year, approximately 1.2 million people worldwide are killed due to traffic accidents where human factors are involved in 90% of them. In order to reduce accidents caused by human error and improve the road safety, automobile companies have been motivated to invest in designing intelligent vehicles and developing driver assistance systems. Therefore, advanced driver assistance system (ADAS) has become an important and active research area in recent years in order to improve driving safety by supporting the driver in critical driving situations that require rapid and safe action.
One of the main modules in ADAS systems is pedestrian detection which aims at detecting and localizing pedestrians at the street and inform drivers about their position and state. I had a chance to work on stereo-based pedestrian detection systems during my PhD research. The aim of the project was to look for pedestrians at the street in different challenging conditions in real time using the video streams captured from two cameras mounted on a car. This area has been studied for several years. However, due to several hardware and software limitations and reliability issues, it was not feasible to use such techniques inside the car. In recent years, deep learning has been a game changer in this area similar to many other domains. With fast development of these techniques, now intelligent vehicles are so close to reality. Pedestrian detection systems are becoming faster and more reliable everyday with great developments in Convolutional Neural Networks which can localize objects in milliseconds and with a very high accuracy.
Reliable pedestrian detection and ADAS systems in general have just been the first step toward a viable and fully autonomous vehicle. With fully development of such vehicles, they are going to affect different aspects of our lives. In the near future, we expect to see a significant transform in transportation, people’s lifestyle, road safety and economy. With self driving cars taking the lead at the roads, there will be a big change in human’s role in transportation. Our next generation may not even have any idea about driver license, road rules and car ownership. Public transportation and delivery systems will be cheaper and more convenient for everyone. One of the most important effects of the autonomous cars will be on saving people’s life. Self driving cars can save around 1.1 million lives in 50 years if they step into the market on 2020 (Figure 1). In economy point of view, self driving cars can provide a lot of financial benefits by saving 1.3 trillion dollars total saving in the US and 5.6 trillion dollars saving in the globe (Figure 2). These savings come from reducing the number of accidents, fuel saving and increased productivity. In addition to that, we can expect a big revenue from the market and a big boost to the job market.
Figure 1 (left): Projected Annual US road death with and without Autonomous Vehicles.
Figure 2 (right): Estimated economic savings from Autonomous Cars.
If you want to know more about Maral and her work, you can follow her on Twitter or connect with her on the Linkedin. In closing, this series would not be able to go on without your participation. If you would like to participate and introduce your work or if you know someone who might be interested please contact me.