SAE Update - March 2024 - 49

A Q&A with SAE instructor
Dr. Hasan Ferdowsi
PROFESSIONAL DEVELOPMENT
any exciting aspects of this research
you're able to share with our readers?
A: In one of my current interdisciplinary
research projects related to AVs, the
primary focus is on enhancing the
intelligence of object detection systems, a
crucial component for AV perception.
Leveraging machine learning, particularly
deep learning techniques like convolutional
and recurrent neural networks, the goal is
to develop algorithms that not only detect
objects but also prioritize specific regions
of interest within the scene. The project
explores the use of region-based attention
mechanisms to identify areas more likely to
contain relevant objects or potential
hazards.
By dynamically allocating computational
resources based on the perceived
importance of different regions, we aim to
optimize the performance of the object
detection system. Ultimately, the research
seeks to improve the safety and reliability
of AVs on the road, with potential
applications in reducing the risk of
accidents and collisions, especially in
complex urban environments. This
collaborative effort involves experts from
diverse fields, including electrical
engineering, robotics, machine learning,
and computer vision, fostering innovation
and contributing to the advancement of
autonomous driving technology for safer
and more efficient transportation systems
in the future. n
By Gretchen Uhrinek, Editor.
UPDATE
March 2024
49
Exploration of Machine Learning and
Neural Networks for ADAS and L4
Vehicle Perception (C2205)
Duration: 2 days
CEUs: 1.3
Format: live online, March 14-15, 2024
In this course, participants will examine an
existing convolutional neural network to
understand how it is constructed and
trained. Using open-source python tools,
participants will create their own
convolutional neural network, train it using a
supplied training data set, and then test your
networks performance. This course is
designed for ADAS and Level 4 development
engineers working on perception using
camera, lidar, and radar sensors, and
validation and test engineers responsible for
functional safety (ISO 26262) and SOTIF
(ISO 21448) where the perception systems
are using AI, neural networks, or other
machine-learning techniques. Policy makers
responsible for regulations regarding public
highway testing of automated vehicles
where perception is based on AI, neural
networks, or other machine learning
techniques will also benefit from this course.
A bachelor's degree in a technical
discipline and familiarity with basic ADAS
functions are recommended prerequisites to
those interested in taking this course.
Familiarity with the python programming
language, or another scripting language
such as visual basic or Matlab, is also
recommended. n
https://www.sae.org/learn/content/C2205/ https://www.sae.org/learn/content/C2205/ https://www.sae.org/learn/content/C2205/
SAE Update - March 2024

Table of Contents for the Digital Edition of SAE Update - March 2024
