Tran Van Manh

Hello, I am

Tran Van Manh

Robotics Engineer  |  Deep RL Researcher  |  Embedded Systems

Download CV View Projects

https://manhbk97.github.io/portfolio/

Who am I?

A Scientist/Roboticist working in Seoul, South Korea

Hi, I'm Tran Van Manh - a robotics engineer with more than 5 years of experience in autonomous navigation systems, motion planning, and AI-driven robotic applications. I enjoy exploring new technologies and tackling challenging engineering problems. Skilled in designing and deploying ROS/ROS2-based systems with behavior-tree decision-making, global and local path planning, and real-time motion control on embedded/edge Linux platforms, with a proven record of transferring reinforcement-learning-based navigation policies from simulation to real-world robots (sim-to-real).

manhtran4321@gmail.com
Seoul, South Korea
Manhbk97
Vietnamese (Native)  |  English β€” TOEIC 740 / IELTS 6.5

Robotics & RL

DRL, ROS, Path Planning, Sim-to-Real

Deep Learning

PyTorch, OpenAI Gym, Policy Networks

Embedded Systems

STM32, Arduino

Control Systems

MPC, Backstepping, PID, Kalman Filter

0 Years of Experience
0 Publications
0 Patent
0 Major Projects
Background

Education & Skills

Education

2022 – 2024

M.Sc. Control and Robot Engineering

Chungbuk National University (CBNU), South Korea

GPA: 4.19 / 4.5

Thesis: Autonomous Navigation in Complex Environments Based on Deep RL Using Selective Policy Switching   [Thesis] [PPT]

2015 – 2020

B.Eng. Control Engineering and Automation

Hanoi University of Science and Technology (HUST), Vietnam

GPA: 3.1 / 4.0

πŸ† Best Graduation Thesis Defense

Skills

Programming
C++CPythonMATLAB
Frameworks & Libraries
ROS / ROS2Nav2PyTorchOpenAI GymOpenCVRTAB-Map
Control
MPPIMPCPIDBacksteppingKalman Filter
Simulation & Tools
GazeboMuJoCoStageV-REPGit / GitHubLaTeX
Embedded
STM32 (ARM)
Work

My Projects

Navigation Β· Planning Β· Behavior Tree September 2024 – Present

Local Planning & Behavior Trees for Service Robots

Developed and deployed local planning and behavior-tree systems for service robots in production restaurant environments, with a focus on real-time performance on edge Linux SBCs.

  • Developed and optimized local planners (TEB, MPC, and MPPI) in Nav2 for autonomous navigation of service and multi-robot systems, reducing on-board computation cost on edge Linux single-board computers (SBCs) by 30% to ensure real-time performance.
  • Implemented full autonomous navigation with Nav2 and designed behavior-tree decision-making with BehaviorTree.CPP for complex scenarios (e.g., narrow-path waiting logic).
  • Developed and maintained the motor controller (emergency-stop logic, robust serial communication, safety handling, battery-voltage filtering and logging); released production software deployed in restaurant environments.
Deploy local planner TEB Testing local path planner TEB Full Autonomous Navigation Behavior Tree Wait Place
RGT Company GitHub Video 1 Video 2 Video 3 Video 4

RGT, South Korea

Deep RL Β· Navigation April 2022 – December 2024

ATC-Mapless Navigation

Developed a deep reinforcement learning framework for autonomous mobile robot mapless navigation in dynamic environments. Includes sim-to-real transfer.

  • Training Deep RL model on simulation environments using GPU.
  • Implement Actor-Critic (A3C) Reinforcement Learning for mobile robots in real environments, SLAM using Rtab-map.
  • Evaluation in static and dynamic environments.
Navigation Environment Distributional RL Robot Platform
ATC Mapless Navigation 1 ATC Mapless Navigation 2
Video 1 Video 2 Paper Certification

Intelligent Robot Lab, CBNU, South Korea

Navigation Β· Hardware August 2020 – February 2022

KIST Sub-project β€” Industrial AMR

Research and development of autonomous mobile robot technology for industrial transportation. Designed navigation framework, kinematic models, and hardware systems for a three-wheel omni robot.

KIST AMR Platform Navigation ROS
KIST AMR Video 1 KIST AMR Video 2
Video 1 Video 2

VKIST, Vietnam

Control Β· Hardware August 2020 – August 2021

PMSM Motor Optimization

Simulation and design of Permanent Magnet Synchronous Motor (PMSM). Designed and simulated magnetic flux distribution for motor optimization.

PMSM Optimization

VKIST, Vietnam

Navigation Β· Hardware Β· AI June 2018 – August 2020

IVASTBot β€” Smart Human-Form Robot

Research, development and embedded programming applying AI for a humanoid robot (phase 01). Designed an intelligent trajectory algorithm for an Omnidirectional Mobile Robot and the corresponding hardware system.

  • Integrate the Navigation stack with Omnidirectional mobile robot.
  • Control PID for 4 Motors of omni wheels.
  • Connect Jetson Tx2 with embedded microcontroller STM32 on ROS.
  • Build simulation environments on Gazebo, Vrep.
4-Wheel Omnidirectional Mobile Robot SLAM Navigation
Testing PID motors Navigation and SLAM
Video 1 Video 2

Physics Institute, VAST, Vietnam

Highlights

News & Achievements

2024

πŸ† Best Paper Award β€” Journal of Korea Robotics Society

"Mapless Navigation with Distributional Reinforcement Learning" received the Best Paper Award from the Journal of Korea Robotics Society, Korea, 2024.

2024

Patent Granted β€” KR Patent MP24-0134/MP24-0141

"Autonomous navigation system in complex environments based on deep RL using selective policy switching for a mobile robot." Co-authored with Prof. Gon-Woo Kim.

View Patent β†’
2024

IEEE Access Publication

"Cooperative Deep Reinforcement Learning Policies for Autonomous Navigation in Complex Environments," IEEE Access, vol. 12, pp. 101053–101065, doi: 10.1109/ACCESS.2024.3429230.

Read Paper β†’
2024

M.Sc. Graduation β€” GPA 4.19/4.5

Completed Master of Science in Control and Robot Engineering at HUST with a thesis on deep RL-based autonomous navigation.

2024

Korean Safety Certification for Human Robot

Evaluated and certified human-robot performance in static and dynamic environments under Korean certification standards.

View Certification β†’
2020

πŸ† Best Graduation Thesis Defense β€” HUST

Thesis "Design intelligent trajectory system for Omnidirectional Mobile Robot" awarded Best Graduation Thesis Defense at HUST.

Conferences

  • The 20th International Conference on Ubiquitous Robots (UR 2023) β€” Hawai'i, USA.
  • The 19th International Conference on Ubiquitous Robots (UR 2022) β€” Jeju, Korea.
  • The 18th Korea Robotics Conference (KRoC 2023) β€” Phoenix Pyeongchang, Gangwon-do.
  • The 17th Korea Robotics Conference (KRoC 2022) β€” Phoenix Pyeongchang, Gangwon-do.
  • 2019 International Conference on Mechatronics, Robotics and Systems Engineering (MoRSE 2019) β€” Bali, Indonesia.
  • Vietnam International Conference and Exhibition on Control and Automation (VCCA 2019) β€” Vietnam.
UR Conference UR 2023 Conference
UR Conference 2 UR 2023Conference
VCCA 2019 VCCA 2019 β€” Vietnam
MoRSE 2019 MoRSE 2019 β€” Bali

Selected Publications

  1. Van Manh Tran and Gon-Woo Kim, "Cooperative Deep Reinforcement Learning Policies for Autonomous Navigation in Complex Environments," IEEE Access, vol. 12, pp. 101053–101065, 2024. [Link]
  2. Van Manh Tran and Gon-Woo Kim, "Mapless Navigation with Distributional Reinforcement Learning," λ‘œλ΄‡ν•™νšŒ 논문지 19.1, pp. 92–97, 2024.
  3. Kim Duyen Ha Thi et al., "Adaptive Control for Uncertain Model of Omni-directional Mobile Robot Based on Radial Basis Function Neural Network," Int. J. Control Autom. Syst. 19, 1715–1727, 2021.
  4. Kim Duyen Ha Thi, Van Manh Tran et al., "Trajectory tracking control for four-wheeled omnidirectional mobile robot using Backstepping aggregated with sliding mode control," ICA-SYMP 2019. [Link]
  5. Nguyen Tung Lam, Van Manh Tran et al., "Nonlinear backstepping-sliding mode control of electro-hydraulic systems," ICERA 2019, Springer, 2020.
  6. Cuong Ng.M, Manh T.V., Duc, D.N., Tung, L.N., Tien, D.P. and Thi, L.T., "Neural network based adaptive control of web transport systems," 2019 International Conference on System Science and Engineering (ICSSE), pp. 124–128. IEEE, 2019.
  7. Ha Thi Kim Duyen, Van Manh Tran et al., "Autonomous Navigation for Omnidirectional Robot Based on Deep Reinforcement Learning," International Journal of Mechanical Engineering and Robotics Research, 2020. (Co-author)
  8. Van Manh Tran et al., "Mapping and Navigation with Four-Wheeled Omnidirectional Mobile Robot Based on Robot Operating System," 2019 International Conference on Mechatronics, Robotics and Systems Engineering (MoRSE), Bali, Indonesia, 2019. (Co-author)
Google Scholar
Get in Touch

Contact

Feel free to reach out for collaborations, job opportunities, or just to chat about robotics!

manhtran4321@gmail.com LinkedIn GitHub Google Scholar