Future of RE
Invited Talk [Future of RE 1 – 3]
Future of RE 1: Nov.11(Fri.)9:20-10:20
Sensing Technologies required for ADAS/AD applications
Kazuoki Matsugatani (Ph. D., DENSO)
ADAS (Advanced Driver Assistance System) and AD (Automated Driving) have received much attention in recent years. Various firms and companies are developing these systems actively.
Vehicle driving task consists of three functions; perception, decision and control. For ADAS, a portion of perception and some control are automated. And for AD, all of functions including decision are automated. Throughout these applications, perception plays an important role, and supporting perception by sensing devices makes driving much safer.
In this talk, firstly, I focus on surround observation sensors used for perception. Typical sensors, camera, radar and LIDAR (Light Detection and Ranging) are introduced and their functions are explained. These components designed for detecting various objects to maintain safe driving.
Then I introduce wireless communication and HMI (Human Machine Interface) devices. Wireless device connects a vehicle to infrastructure, and HMI connects vehicle to human driver. For wireless communication, local or ad-Hoc media of vehicle to infrastructure and inter vehicle communication are utilized. And cellular based radio is also used for vehicles. Regarding HMI devices, DSM (Driver Status Monitor) and HUD (Head Up Display) are typical components required for ADAS/AD.
Finally, I show our recent activities for investing and demonstrating ADAS/AD applications. Our ultimate objective is ‘ZERO ACCIDENTS’, and to make driving ‘FUN’. We have been challenging persistently to realize this objective.
Dr. Matsugatani has more than 25 years experience of R&D activities in electric engineering, including semiconductor physics, microwave and millimeter-wave circuits, wireless communications and ADAS (Advanced Driver Assistance System).
After receiving master degree from Kyoto University in 1989, he joined DENSO CORPRATION. He has been involved in developing high frequency transistors and monolithic microwave / millimeter-wave integrated circuits (MMIC) for automotive radar. After that, he has also developed and designed wireless radio circuits, antennas, communication protocols and system configurations for road-to-vehicle and inter-vehicle communications. In 2010, he has received Ph.D on “Study of Microwave Planar Antennas Using Periodic Structures” from Nagoya Institute of Technology.
In 2010, Dr. Matsugatani had served as Director of Corporate R&D Division 3, DENSO CORPORATION. And in 2015 he had moved to R&D Division 1, then in 2016, he has established ADAS Business and Technology Development Division and focusing on ADAS and AD (Automated Driving) development as the Director.
Future of RE 2: Nov.11(Fri.)13:30-15:30
Machine Learning as a Programming Paradigm and its Implications to Requirements Engineering
Hiroshi Maruyama (Chief Strategy Officer, Preferred Networks, Inc.)
Application areas of machine learning is quickly expanding, from image recognition to language translation, game playing, and autonomous driving. Machine learning can be viewed as a tool for building a system inductively from a set of input-output examples, where specifications of such a system are given as training data sets. How should we translate our requirements into these specifications, that is, how should be prepare appropriate training data sets that reasonably represent the given requirements? Since new training data sets become continuously available for online systems, the specifications also continuously change over time. How should we assure all these specifications are consistent with the requirements?
There are a number of open questions with this new programming paradigm. This talk will explore some of these open questions and discuss their implica-tions and the opportunities to the requirements engineering community.
Dr. Hiroshi Maruyama has spent 26 years in IBM Research, Tokyo Research Laboratory, working on various computer science areas such as artificial intelligence, natural language processing, machine translation, hand-writing recognition, multimedia, XML, Web Services, and security.
He was the director of IBM Tokyo Research Laboratory from 2006 to 2009. From 2011 to 2016, he was a professor at the Institute of Statistical Mathematics where he worked on two projects — Systems Resilience and Data Scientist Training Network, among other research activities related to statistics and computer science.He joined Preferred Networks, Inc. in April 2016 as the chief strategy officer. His research interests include machine learning, cyber-physical systems, and computer science in general.
Future of RE 3: Nov.12(Sat.)13:30-15:00