I am currently working at Ford Motor Company as a Research Scientist on the Autonomous Driving project as a part of Ford's Research and Innovation Center.
I have a PhD in Electrical and Computer Engineering at
Carnegie Mellon University (CMU), 2015.
I obtained my M.S degree from ECE department of CMU, 2011, and my Bachelor’s degree is in
Communication Systems (System de communication) from École Polytechnique
Fédérale de Lausanne (EPFL) in Switzerland, 2009.
I have years of research and system engineering experience in wireless communications, specializing in safety applications using vehicular networks and DSRC. Intensive experience in both simulation and emulation of cyber-physical systems, distributed systems, sensor networks, vehicle communications and strong programming skills. Exceptional interpersonal, leadership and communication skills with a dedication to promoting effective teamwork.
August 2009 - May 2015
Electrical and Computer Engineering
Thesis Statement: A fusion of vehicular networks and vehicle-resident sensing enables co-operative driving among autonomous and manual vehicles, leading to safety and high throughput at intersections.
August 2009 - May 2011
Electrical and Computer Engineering
ABD
August 2004 - May 2008
Communication Systems
ABD
August 2009 - Present
Designed and developed new vehicular networks protocols using Vehicle-to-Vehicle (V2V) communications to enable co-operative driving in the context of autonomous vehicles.
Designed and developed methods to enable the safe co-existence of manual and autonomous vehicles at intersections.
Designed and developed the hybrid emulator-simulator for vehicular networks, called AutoSim
August 2009 - Present
Designed and developed active safety applications using vehicular communications (V2X)
Prototyped and implemented the proposed protocols on CMU-GM autonomous vehicle
Summer 2014
Systems Group, Manager: Peerapol, Tinnakornsrisuphap
Performance Improvement of Video Telephony over WLAN
Summer 2013
System Integration Group, Manager: Michael Dimare, Mentor: Vito Salluce
System performance evaluation of LTE-D: Designing test cases, deployment, evaluation and analysis
802.11 n/p: Throughput evaluation of Wi-Fi/DSRC on Kingfisher
Fall 2014
ABD
Spring 2012 and Spring 2014
ABD
Spring 2008
ABD
Fall 2007 and Spring 2008
ABD
Designed a family of distributed intersection management protocols, STIP (Spatio-Temporal Intersection Protocols), to provide safe and efficient traverse of vehicles through road intersections and roundabouts. STIP incorporate vehicular networks to enable co-operative driving of manual and autonomous vehicles. We analyze potential deadlock and starvation conditions which can affect our distributed cyber-physical system. We present a mathematical proof for the deadlock-freedom and liveliness of our proposed system.
To incorporate DSRC/WAVE in the STIP framework, we implement realistic communication models. Additionally, we study the impact of imperfect communication on our V2V-intersection protocols and leverage the use of realistic DSRC channel propagation models such as the Nakagami-m model. Additionally, we analyze the communication reliability of our proposed active safety applications.
Localization and positioning accuracy play an important role in safety applications. We study the effects of position inaccuracy on our STIP framework by implementing realistic GPS models. We design and implement a method to deal with position inaccuracies to guarantee the safety of our intersection protocols.
We design a new synchronization-based method to manage the synchronous and continuous arrival and passage of vehicles at intersection. We analyze the benefits of this scheme in maximizing the usage of intersection capacity. This method can be also beneficial for a wide range of non-vehicular applications.
We design and implement new protocols for managing the mixed traffic of human-driven and autonomous vehicles through intersections. Leveraging V2V, V2I and on-board sensor systems, we design communication-based and perception-based protocols which enable the safe co-existence of manually-driven and autonomous vehicles. These methods are evaluated to demonstrate the improvements obtained in safety and throughput at intersections.
To evaluate our Spatio-temporal Intersection Protocols (STIP), we use our model-based emulator-simulator, AutoSim. This simulator is the next generation of GrooveNet. This tool is used to implement and evaluate our STIP framework. AutoSim provides a hybrid emulation and simulation environment for vehicular communications. The communication interfaces for DSRC communication as well as on-board sensory interfaces is implemented to enable real cars instrumented with DSRC to react in real-time with simulated vehicles. We also support modeling of different aspects of mobility protocols. The architecture consists of several models such as the Control, Communication, Mobility, Localization and Path Tracking. AutoSim also support modeling of Vehicle-to-Vehicle (V2V) and Vehicle-to-Infrastructure (V2I) communications.
Our STIP methods are being implemented on the autonomous vehicle platform.