PhD/Postdoc Positions
We are always looking for highly-motivated candidates who are
interested in developing intelligent vision and control techniques
to enhance the intelligence and autonomy of robots including UAVs.
- Topics:
We are interested in topics that are
essential to enhancing the intelligence of robots, which include
but are not limited to:
Deep Learning and Deep Reinforcement Learning-based Navigation for
robots and UAVs
- Deep learning-based scene understanding, object detection and
tracking, Large Language Models (LLM)
- Deep reinforcement learning-based control and navigation.
Vision-based navigation and control for Robotics
- Vision-based motion planning and obstacle avoidance
- Vision-based object detection and tracking
Swarm robotics
-
Cooperative SLAM, cooperative control and multi-robot obstacle
avoidance for a swarm of robots.
- Multi-agent reinforcement learning
Visual Odometry and SLAM for robots
- Event-based, vision-based or lidar-based visual odometry
- Event-based, vision-based or lidar-based mapping
Soft and continuum robots
- Dynamic control and grasping of soft/continuum arms
- Requirements:
The candidate is expected to have a
Master's Degree for applying for a PhD or
Doctor's Degree for applying for a
postdoc position. He/she should have a relevant research background (Learning, UAV,
robotics) and programming skills with C/C++, python and ROS. Experience
with drone/robotics development skills such as ROS/Gazebo, OpenCV, and PX4/Issac
Gym is a plus.
- How
to apply:
Please email your CV, a PPT file that
describes your research experience in detail, and academic
transcripts to Dr. Peng Lu (lupeng@hku.hk),
quoting [PhD/Postdoc Application]+[your name] in the email subject.