Abstract: In emerging autonomous and semi-autonomous vehicles, accurate environmental perception by automotive cyber-physical platforms is critical for achieving safety and driving performance goals. An efficient perception solution capable of high fidelity environment modeling can improve Advanced Driver Assistance System (ADAS) performance and reduce the number of lives lost to traffic accidents as a result of human driving errors. Enabling robust perception for vehicles with ADAS requires solving multiple complex problems related to the selection and placement of sensors, object detection, and sensor fusion. Current methods address these problems in isolation, which leads to inefficient solutions. In this thesis, we present a novel perception architecture exploration framework for automotive cyber-physical platforms capable of global co-optimization of deep learning and sensing infrastructure.

Adviser: Dr. Sudeep Pasricha
Co-Adviser: N/A
Non-ECE Member: Dr. Bret Windom
Member 3: Dr. Anura Jayasumana
Addional Members: N/A

Publications:
Joydeep Dey, Wes Taylor and Sudeep Pasricha, "VESPA: A Framework for Optimizing Heterogeneous Sensor Placement and Orientation for Autonomous Vehicles,", IEEE Consumer Electronics Magazine, vol. 10, no. 2, pp. 16-26, 2021

Joydeep Dey, Sudeep Pasricha, “PASTA: Perception architecture search technique for Advanced Driver Assistance Systems”, Design Automation Conference (DAC), 2022

Program of Study:
ECE 554
ECE 561
ECE 799
ECE 441
CS 545
CS 440
GRAD 550
STAT 581A3