About the role
AI summarisedThe Cognitive and Human-like AI group at A*STAR's Institute of High Performance Computing is seeking a PhD-level scientist to conduct R&D in Embodied Artificial Intelligence (EAI) for human-robot collaboration in manufacturing and trade. The role involves designing algorithms for multimodal perception, spatial awareness, and intent prediction, as well as developing and integrating human-robot interaction modules.
ResearchFull-time
Key Responsibilities
- Research in human-robot interaction involving multimodal perception, human and situation understanding, world representation, decision making, learning and behaviour.
- Design EAI algorithms such as for human-centric spatial awareness and intent prediction to ensure safe, fluid, and proactive collaboration between humans and robots in shared workspaces.
- Development, experimentation, testing and system integration of human-robot interaction modules for various real world robotics applications.
- Actively collaborate with other researchers in R&D activities including scientific writing, algorithm development, testing and deployment of the intelligence systems.
Requirements
- PhD in Robotics or a field of study related to EAI (e.g. Computer Vision, Natural Language Processing, Machine Learning, Cognitive Science, Cognitive Architectures, Robotic Foundation Models, etc.)
- Good programming skills in python and/or C++ is essential.
- Experience with ROS, Linux and/or other robotic middleware system is an advantage.
- Experience in dataset curation and fine-tuning Vision-Language-Action (VLA) or Vision-Language Models (VLM) for robotics is an advantage.
- Past work in the following areas is an advantage: 3D vision and spatial reasoning; human-robot interaction; situation representation of human-agent shared tasks/spaces; representation, modelling and reasoning of human actions/behaviours.
- Good communication skill, a team player and resourceful.