Abdulrahman Al-Shanoon, Ph.D.

Hi there! I'm an AI/ML Scientist at General Motors of Canada, where I apply my expertise in artificial intelligence, machine learning, autonomous robotics, computer vision, deep learning, reinforcement learning, and perception to innovate in advanced driver assistance system (ADAS) mapping and software development. I work with the Software Defined Vehicle (SDV) team and the Mapping Software group to create and implement machine learning models for GM Map Management System (MMS), which is used by Super Cruise and Ultra Cruise programs.


Formal Bio     Google Scholar     LinkedIn     YouTube     ORCID

aalshanoon at gmail dot com

📍 Toronto, ON, Canada
Some highlights from our research:

Smarter Robots: Learning to Make Sequential Decisions

Sequential decision-making issue is a challenging task in skilled robots. This paper takes a step towards solving the issue by introducing a self-learning strategy to manipulate unknown objects in challenging scenarios based on minimal prior knowledge. The developed system learns jointly pre-grasping and grasping actions using model-free deep reinforcement learning for robotic manipulation task.

From Choice to Action: Sequential Decisions in Robotics: PDF

Grasping Unknown Objects: Robots Learning New Skills

Grasping unfamiliar objects (unknown during training) based on limited prior knowledge is a challenging task in robotic manipulation. We introduce a robotic grasping strategy based on the model-free deep reinforcement learning, named Deep Reinforcement Grasp Policy. The developed system demands minimal training time and limited simple objects in simulation and generalizes efficiently on novel objects in real-world scenario.

Beyond the Familiar: Robots Learning to Grasp Anything: PDF

Autonomous (8-DOF) Mobile Manipulators

The fourth industrial revolution (industry 4.0) demands high-autonomy and intelligence robotic manipulators. This work sets out to develop an autonomous 3D Visual Servoing system based on DeepNet, implemented in a sophisticated mobile manipulator system, and utilized single RGB image. Two main steps construct the entire system: first, perception network to estimate the pose of objects in 3D space. Second, the pose estimation data was then used in a 3D visual servoing scheme to control the motion of AMM system.

Industry 4.0 Calls for Highly Autonomous and Intelligent Manipulators: PDF

3D Visual Servoing and Deep Neural Networks for Robotic Manipulation

The critical challenge, for robot–object-interaction, is to estimate visually the pose of the target object in a 3D space and combine it into a vision-based control scheme in manipulation applications. We proposes a novel reliable framework for deep ConvNet combined with visual servoing using a single RGB camera. We introduce an extensive system called Deep-Visual-Servoing (DVS) that addresses an integration of: (I) training of deep-CNNs using synthetic dataset only and operates successfully in real-world scenario, (II) continuous 3 D object pose estimation as the sensing feedback in a 3D visual servoing control scheme.

Deep-Visual-Servoing for Robotic Manipulation: PDF

Position-Based Visual Servoing for Mobile Robot

Position-based visual servoing control system that is implemented in a Clearpath Husky mobile robot. The vision sensor used is a Kinect V1 by Microsoft. The results are organized in to two separate tests; Straight and Steering. Both tests demonstrated the control systems ability to regulate the robots position and orientation to the desired values.

Vision-Based Mobile Robot Control: PDF

Autonomous Navigation and Self-Charging for Mobile Robot

A Husky Mobile Robot autonomously navigates to a charging station using a laser sensor and a pre-defined map to move while avoiding new and unexpected obstacles. A single camera then helps refine its approach, ensuring precise docking at the charging station. This technology enhances industrial robots by enabling self-charging capabilities.

Robot Self-charging and path planning technology: PDF

Patent:

  • • Wind Monitoring System for A Vehicle. P108471-PRI-NP-US01. 11/2024.

  • • Tandem Electric Vehicle Charging. Google Patents. P104086-PRI-NP-US01. 10/2024.

  • • Vehicle systems and methods for identification and deduplication of roadway objects P108591-PRI-NP-US01. 09/2024.

  • • Using Vehicle Sensor Inputs to Control Active Aero Devices. P105910-PRI-NP-US01. 10/2023.

Research Publication:

  • Tan, Aaron H., Al-Shanoon Abdulrahman, Haoxiang Lang, and Ying Wang. "MOBILE ROBOT DOCKING WITH OBSTACLE AVOIDANCE AND VISUAL SERVOING." International Journal of Robotics and Automation 38, no. 2 (2023).
  • Al-Shanoon Abdulrahman, and Haoxiang Lang. "Learn to Grasp Unknown-Adjacent Objects for Sequential Robotic Manipulation." Journal of Intelligent & Robotic Systems 105, no. 4 (2022): 83.
  • Al-Shanoon, Abdulrahman, Yanjun Wang, and Haoxiang Lang. "DeepNet-Based 3D Visual Servoing Robotic Manipulation." Journal of Sensors 2022 (2022).
  • Al-Shanoon, Abdulrahman, and Haoxiang Lang. "Robotic manipulation based on 3-D visual servoing and deep neural networks." Robotics and Autonomous Systems 152 (2022): 104041.
  • Al-Shanoon, Abdulrahman, and Haoxiang Lang. "Vision-Based Hand Gesture Recognition With Deep Machine Learning for Visual Servoing." In International Design Engineering Technical Conferences and Computers and Information in Engineering Conference, vol. 51814, p. V05BT07A050. American Society of Mechanical Engineers, 2018.
  • Al-Shanoon, Abdulrahman, Haoxiang Lang, and Ying Wang. "Learn to grasp unknown objects in robotic manipulation." Intelligent Service Robotics 14, no. 4 (2021): 571.
  • Al-Shanoon, Abdulrahman. "Developing a mobile manipulation system to handle unknown and unstructured objects." PhD diss., 2021.
  • Al-Shanoon, Abdulrahman, Aaron Hao Tan, Haoxiang Lang, and Ying Wang. "Mobile Robot Regulation with Position Based Visual Servoing." In 2018 IEEE International Conference on Computational Intelligence and Virtual Environments for Measurement Systems and Applications (CIVEMSA), pp. 1-6. IEEE, 2018.
  • Tan, Aaron, Al-Shanoon Abdulrahman, Haoxiang Lang, and Moustafa El-Gindy. "Mobile Robot Regulation With Image Based Visual Servoing." In International Design Engineering Technical Conferences and Computers and Information in Engineering Conference, vol. 51807, p. V05AT07A078. American Society of Mechanical Engineers, 2018.
  • Al-Shanoon, Abdulrahman, Siti Anom Ahmad, Mohd Hassan, and B. Khair. "Gripping an Object Based on Inspection of Slip Events for a Robotic Hand Model." Advanced Science Letters 23, no. 6 (2017): 5133-5137.
  • Al-Shanoon, Abdulrahman, Siti Anom Ahmad, and Mohd Khair B. Hassan. "Re-gripping analysis based on implementation of slip-detection device for robotic hand model." In 2016 IEEE Region 10 Symposium (TENSYMP), pp. 203-206. IEEE, 2016.
  • Al-Shanoon, Abdulrahman, Siti Anom Ahmad, and Mohd Khair B. Hassan. "Investigation of piezoresistive sensor for robotic gripping operations." In 2015 IEEE Student Symposium in Biomedical Engineering & Sciences (ISSBES), pp. 54-58. IEEE, 2015.
  • Al-Shanoon, Abdulrahman, and Siti Anom Ahmad. "Slip detection with accelerometer and tactile sensors in a robotic hand model." In IOP Conference Series: Materials Science and Engineering, vol. 99, no. 1, p. 012001. IOP Publishing, 2015.
  • Al-Shanoon, Abdulrahman, Siti Anom Ahmad, and Mohd Khair B. Hassan. "A susceptibility study on piezoresistive sensor in a pliable and rigid robotic claws model." (2015): 9106-Q111.
  • Al-Shanoon, Abdulrahman. "Gripping controller design for a one-degree-of-freedom robotic hand model based on slip detection." M.Sc thesis, 2016.

Teaching Experience:

Teaching Assistant, Faculty of Engineering, Ontario Tech University, ON, Canada 2017-2021

  • Taught a wide range of engineering subjects: “Dynamics, Introduction to Engineering, Robotics and Automation, Microprocessors and Digital Systems, Electromechanical Energy Conversion.”
  • Contributed to the creation of methods and techniques for online teaching during the lockdown, which have now become the standard for virtual education practices.
  • Provided instruction in diverse university-level classes.
  • Oversaw the curriculum planning for undergraduate subjects.
  • Graded assignments, midterms, and final exams for multiple courses.
  • Developed class tutorials, lab materials, assignments, and exam questions.
  • Held weekly consultation hours for students. Participated in weekly review meetings with the department supervisor.
  • Formulated grading rubrics for coursework and examinations.

Laboratory Director at Dijlah University College, Iraq 2011-2013

  • In charge of Laboratory Official Digital Electronics and Software System.
  • Teaching digital electronics and computer subjects to university level students.
  • Compiled a project report with forward-looking recommendations and findings.
  • Developed assignment modules for various chapters.
  • Evaluated student comprehension and employed tailored teaching methods.
  • Guided students through intricate numerical challenges. Marked students' exam papers.
  • Guided students through the curriculum in preparation for their final exams.
  • Monitored and adhered to the anticipated curriculum timelines.
  • Fostered an engaging learning atmosphere through passion and enthusiasm.
  • Acquired insights into the integration of educational theory and practical application.

Education:

- Doctor of Philosophy (Ph.D.) - AI and Advanced Robotics. Ontario, Canada 2017-2021
Faculty of Engineering, Ontario Tech University

- Master of Science (M.Sc.) - Robotics Engineering. Malaysia 2014-2016
Faculty of Engineering, Universiti Putra Malaysia

- Bachelor of Engineering (B.Eng.) Honors: First Rank. Iraq 2007-2011
Electrical & Electronics Engineering, Middle Technical University

  • Graduated with honors - First rank.

Work Experience:

AI/ML Scientist, General Motors of Canada. ON, Canada

2021-Present

  • Developer of AI/ML innovation projects in ADAS features, crafting ML models for autonomous driving technologies. Played a key role within the Mapping Software group, supporting Software Defined Vehicle (SDV) team and driving forward technological advancements.
  • Led entire process of training, validating, and registering ML models for ADAS perception capabilities.
  • Led projects focused on Active Safety Features, proactively preventing collisions and saving lives by proposing and executing innovative engineering solutions in advanced future vehicles.
  • Innovated as an AI/ML Developer, significantly advancing projects in Active Safety Features, Perception, Computer Vision, and ML models, thereby enhancing vehicle safety and trailering technology.

    • Postdoctoral Research Fellow, Ontario Tech University. ON, Canada

    04/2021-10/2021

  • Project Scope: Advanced Robotic Motion Control, Machine Learning, and Computer Vision.
  • Develop & implement a Machine Learning model for Path Planning & Obstacle Avoidance, and Self-Learning Control Strategy for Manipulator Robots.
  • Using Computer Vision and Reinforcement Learning methods, developing & implementing an advanced Machine Learning model to accomplish proficient Robotic Manipulation and object grasping.
  • Validate and analyse the knowledge learned during training sessions, then execute it in real-life challenging situations.

    • Teaching Assistant, Ontario Tech University. ON, Canada

    2017-2021

  • Taught a range of engineering subjects: “AI applications, Dynamics, Introduction to Engineering, AI methods for Robotics Applications, Robotics and Automation, Microprocessors and Digital Systems, Electromechanical Energy Conversion.”

    • Mitacs Accelerate Internship. ON, Canada

    2019-2020

  • Partner Organization: Senturing Technologies Ltd and Ontario Tech University.
  • Project Title: Safe and Low-cost Robot Grasping through Impedance Control and Deep Learning.
  • Proposed, implemented, and verified a Vision-Based Deep ML 3D Object Estimation Model combined with Advanced Motion Control System for a sophisticated Long-Range Mobile-Manipulator Robot.
  • Introduced, implemented, and validated a novel Deep ML framework using Visual Perception Network for skilled Robotic Control System and Self-Learning Strategies.
  • The proposed Advanced systems included Simulation settings & Real-world Implementations.

    • Laboratory Director, Dijlah University College

    2011-2013

  • In charge of Laboratory Official Digital Electronics and Software System.
  • Teaching digital electronics and computer subjects to university level students.
  • Developed assignment modules for various chapters.