Your daily schedule will include frequent interactions with members of a multidisciplinary team.
- You are comfortable working in sea trials and pool trials, instead of a normal cabin in the office.
- You build safe and scalable robotic systems with clean and documented code.
The scope of the job
The scope of this position is in the development of intervention (or underwater manipulation) for an autonomous underwater vehicle capable of flying over SS3 – SS4 swell conditions and currents of up to 4 knots for stability in offshore conditions. At the most basic level, you will develop the AUV maneuvers required for non-destructive testing of 1 to 3 degrees of freedom coupled to multiple actuators and linear mechanisms and taking into account the forces and force feedback needed to ensure that the task is complete. In the most advanced case, you will develop 5-7 degrees of freedom manipulator path following for the same task but now on a sloped surface. These tasks are all part of the requirement of an AUV capable of hovering to perform repair and maintenance inspection work on offshore structures. You will need to connect with other remote response, perception and operations groups and test the technology within the ocean basin of the Technology Center for Offshore Marine Singapore (TCOMS) and in the open waters and rapid currents of Singapore. .
- Research and implement perception and fusion algorithms using camera and sonar.
- Research and implement control strategies for multi DOF robotic manipulators.
- Implement inverse kinematics and path planning algorithms for multi-degree-of-freedom manipulators.
- Exploration of new methods such as reinforcement learning where necessary to ensure high pass rate and repeatability. Atomic actions are not part of the equation, the key is to merge perception with path planning on a moving platform.
- Integration of the robotic stack with the software architecture.
- Testing and debugging robotic algorithms in loop simulation software
- MSc in Computer Engineering / Mechatronics / Mechanical Engineering (robotics equivalent)
- Experience with 5-7 degrees of freedom robot manipulator control and strategies
- Experience with perception and fusion algorithms and libraries including OpenCV, Pointcloud library, tracking filters, computer vision techniques
- Experience with AI, deep learning and machine learning algorithms such as YOLO and Faster RCNN
- Experience with obstacle collision avoidance trajectory planning algorithms such as A * STAR, PRM, etc.
- Knowledge of inverse kinematics.
- Experience of reinforcement learning is a plus.
- Proficient in Python and C ++
- Proficient in Linux
Site: Kent Ridge Campus
department : Mechanical Engineering
Eligible for employee recommendation: No
Job Application ID: 6920