Internship

05/2022-08/2022

Handheld Controller for Teleoperation

​

Project Description: 

With the advantages of being compact and lightweight, handheld controllers are attractive tools for teleoperation. Without mechanical linkages, the physical footprint can be significantly reduced. Besides, an intuitive ungrounded design can assist the operators to manipulate the robot through teleoperation in a more natural way without motion constraints. This project aims to develop a handheld controller for remotely controlling a robot. A 9-axis IMU will be adopted to measure the finger gripper’s gesture of the controller and a depth camera will be applied to track its 3d position. The handheld controller will be first tested in a simulated environment and then evaluated on a physical robotic platform.

​

Essential requirements: Python

Desirable requirements: Signal Processin

05/2022-08/2022

Machine Learning-Based Human-Robot Shared Control

​

Project Description: 

In most of the current robotic systems, teleoperation dominates the mainstream and is the major form of control. However, most of the tasks contain a lot of relatively simple but repetitive sub-tasks, which brings substantial burdens to the operator.

Therefore, it is necessary to bring autonomy to current robotic systems. In this case, we aim to build a machine learning-based human-robot shared control framework, which can automate some of the sub-tasks via learning from demonstration.

​

Essential requirements: Python programming skills

Desirable requirements: Vision, Machine Learning

05/2022-08/2022

Surgical Tool Segmentation and Localization

​

Project Description: 

In robotic surgery, tool tracking is important. Despite recent advances in tool tracking, existing approaches may suffer from issues such as occlusions, shadows, motion blur or specular reflections. In this project, real-time surgical tool tracking, segmentation and localization based on deep learning techniques will be explored. The surgical tool will be augmented with feature enhancement to provide effective visual feedback to the operator during teleoperation.

​

Essential requirements: Python programming skills

Desirable requirements: Vision, Deep learning

​

05/2022-08/2022

Workspace Analysis for Parallel Robot or Continuum Robot

​

Project Description: 

Workspace analysis is essential for robotic manipulators, which helps researchers study, evaluate, and optimize their designs based on specific criteria with due consideration of ergonomics and usability. Although workspace analysis is a common research topic, current solutions provide design-specific evaluation, and there is a lack of generic software tools for different hardware configurations. WSRender, a versatile research-oriented framework for workspace analysis and visualization has been proposed. The limitation of the current WSRender version is the lack of a calculation module for parallel robots or continuum robots. Therefore, this project aims to include workspace analysis for parallel robots or continuum robots.

​

Essential requirements: Matlab

Desirable requirements: Robotics