Author Image

Hi, I am Ramin

Ramin Nabati

Autonomous Vehicle Sensor Fusion Engineer at Ford Motor Company

I’m Ramin, currently working with the Autonomy perception team at Ford, developing perception solutions for autonomous driving. Prior to joining Ford, I was a Ph.D. student at the University of Tennessee Knoxville (UTK) in the Advanced Imaging and Collaborative Information Processing (AICIP) lab advised by Dr. Hairong Qi, where my research was focused on radar-camera sensor fusion for object detection and tracking in autonomous vehicles. During my time at UTK, I also led the Advanced Driver Assistance Systems (ADAS) and Connected and Automated Vehicles (CAV) teams in the EcoCAR 3 and EcoCAR Mobility Challenge competitions.

Experiences

1
Autonomous Vehicle Sensor Fusion Engineer
Ford Motor Company

Aug 2021 - Present, Ann Arbor, MI

Ford Autonomous Vehicles (AV) & Mobility is dedicated to enhancing freedom of movement through a suite of mobility products and services, from micromobility to microtransit.

Responsibilities:
  • Research, design, develop and deploy perception algorithms for autonomous vehicles, including object detection, object tracking and sensor fusion algorithms.

Sensor Fusion Intern
Blueberry Technology

June 2020 - Aug 2020, San Jose, CA

Blueberry Technology enhances user mobility by providing autonomous wheelchairs for indoor use.

Responsibilities:
  • Implementing and optimizing a real-time monocular depth estimation algorithm for an autonomous indoor wheelchair. Applying sensor fusion and optimizing the model for the target hardware architecture to achieve real-time performance.
2

3
Connected and Automated Vehicle (CAV) Team Lead and GRA
EcoCAR Mobility Challenge - UTK

August 2018 - Aug 2021, Knoxville, TN

The EcoCAR Mobility Challenge was the 12th U.S. Department of Energy (DOE) Advanced Vehicle Technology Competition (AVTC) series. The four-year competition challenged 11 university teams to apply advanced propulsion systems, as well as connected and automated vehicle technology to improve the energy efficiency, safety and consumer appeal of the 2019 Chevrolet Blazer.

Responsibilities:
  • Leading the UTK Connected and Automated Vehicle technologies (CAVs) team in the EcoCAR Mobility Challenge competition. With a focus on SAE level 2 autonomy, the CAVs team designs the onboard perception system, Vehicle-to-X communication system and lateral/longitudinal autonomy system for a 2019 Chevrolet Blazer.

Advanced Driver Assistance Systems (ADAS) Team Lead and GRA
EcoCAR 3 - UTK

August 2017 - May 2018, Knoxville, TN

EcoCAR 3 was the 11th U.S. Department of Energy (DOE) Advanced Vehicle Technology Competition (AVTC) series and is North America’s premier collegiate automotive engineering competition. The U.S. DOE and General Motors are challenged 16 North American universities to redesign a Chevrolet Camaro into a hybrid-electric car to reduce environmental impact, while maintaining the muscle and performance expected from this iconic American car.

Responsibilities:
  • Leading the UTK Advanced Driver Assistance System (ADAS) team in the EcoCAR3 competition. The ADAS team focuses on integrating sensing system on a 2016 Chevrolet Camaro and using this system to deploy driver feedback to improve efficiency and safety.
4

Education

Projects

CenterFusion
Lead Author 2021

A center-based radar and camera fusion for 3D object detection in autonomous vehicles.

EcoCAR Mobility Challenge
CAV Team Lead Aug 2018 - Aug 2021

12th U.S. Department of Energy (DOE) Advanced Vehicle Technology Competition (AVTC) series, challenging 11 university teams to apply advanced propulsion systems, as well as connected and automated vehicle technology to improve the energy efficiency, safety and consumer appeal of the 2019 Chevrolet Blazer.

EcoCAR 3
ADAS Team Lead Aug 2017 - May 2018

EcoCAR3 is challenging 16 North American university teams to redesign a 2016 Chevrolet Camaro. The ADAS team focuses on integrating the sensing system on the Camaro and deploy driver feedback to improve efficiency and safety.

IARPA fMoW Challenge
IARPA fMoW Challenge
Contributor 2017

The IARPA Functional Map of the World (fMoW) challenge focuses on promoting research in object identification and classification to automatically identify facility, building, and land use from satellite imagery. The dataset consists of 4- and 8-band multispectral images in 63 categories.

Spacenet 3: Road Network Detection
Spacenet 3: Road Network Detection
Contributor 2018

The Spacenet 3 challenge is focused on determining road networks and routing information directly from satellite imagery. The SpaceNet 3 Dataset contains ~8,000 km of roads across the four SpaceNet Areas of Interest.

Publications

CenterFusion: Center-based Radar and Camera Fusion for 3D Object Detection

In this paper, we focus on the problem of radar and camera sensor fusion and propose a middle-fusion approach to exploit both radar and camera data for 3D object detection. Our approach, called CenterFusion, first uses a center point detection network to detect objects by identifying their center points on the image. It then solves the key data association problem using a novel frustum-based method to associate the radar detections to their corresponding object’s center point. The associated radar detections are used to generate radar-based feature maps to complement the image features, and regress to object properties such as depth, rotation and velocity.

CFTrack: Center-based Radar and Camera Fusion for 3D Multi-Object Tracking

In this work, we propose an end-to-end network for joint object detection and tracking based on radar and camera sensor fusion. Our proposed method uses a center-based radar-camera fusion algorithm for object detection and utilizes a greedy algorithm for object association.

Radar-Camera Sensor Fusion for Joint Object Detection and Distance Estimation in Autonomous Vehicles

In this paper we present a novel radar-camera sensor fusion framework for accurate object detection and distance estimation in autonomous driving scenarios. The proposed architecture uses a middle-fusion approach to fuse the radar point clouds and RGB images.

RRPN: Radar Region Proposal Network for Object Detection in Autonomous Vehicles

In this paper we introduce RRPN, a Radar-based real-time region proposal algorithm for object detection in autonomous driving vehicles. RRPN generates object proposals by mapping Radar detections to the image coordinate system and generating pre-defined anchor boxes for each mapped Radar detection point. These anchor boxes are then transformed and scaled based on the object’s distance from the vehicle, to provide more accurate proposals for the detected objects.

Recent Posts