Research Fellow, Perception
2 days left
- Full Time
The research scope for this position is in perception and navigation development for an Autonomous Underwater Vehicle operating in offshore environment. We will be developing advanced techniques for structural tracking and methods for perception generalisation across multiple types of offshore infrastructure. Perception and navigation strategies around complex jackets will be devised with a high level of scale as the focus for the techniques utilised. The techniques will start first with a mapping step (not necessarily used for localisation) followed by perception and navigation step. The development will be first deployed in the digital twin model of the Technology Centre for Offshore Marine Singapore (TCOMS) ocean basin before being tested in the open waters of Singapore.
- Research and implement perception and fusion algorithms utilising sensors such as camera, stereo camera and sonar.
- Explore cost map and mapping techniques to achieve a certain level of mapping of the environment in 3D space. Be able to extend this is to enable global localisation against this pre-existing a-prior map that is built from previous mapping run.
- Research and explore possible localisation algorithms fusing multiple navigation and sensor sources with SLAM for handling of edge cases in vehicle navigation.
- Implement optimal path planning strategies to achieve efficient inspection coverage according to CSWIP 3.3U/3.4U Underwater Inspection standards.
- Integration of robotics stack with the software architecture
- Test and debug of robotics algorithms in software in the loop simulation
- PhD in Computer Engineering or Computer Science
- Research or industry experience writing code for complex robotic systems
- Experience with Robot Operating System framework
- 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 SLAM for localisation methods
- Experience with path planning algorithms such as A*STAR
- Proficient in Python and C++
- Proficient in Linux
Location: Kent Ridge Campus
Department: Mechanical Engineering
Employee Referral Eligible: No
Job requisition ID: 6921