Qualification Type: | PhD |
---|---|
Location: | Manchester |
Funding for: | UK Students |
Funding amount: | The funding is for home students and will cover tuition fees and provide a tax free stipend set at the UKRI rate (£18,622 for 2023/24). |
Hours: | Full Time |
Placed On: | 30th January 2024 |
---|---|
Closes: | 31st March 2024 |
This 3.5 year PhD project is funded by The Department of Electrical and Electrical Engineering. The funding is for home students and will cover tuition fees and provide a tax free stipend set at the UKRI rate (£18,622 for 2023/24).
In the realm of human-robot collaboration, the precise recognition of human intentions is pivotal for robots to understand user needs and offer natural, effective assistance. This task, however, presents substantial challenges in real-time operations, especially in complex scenarios requiring quick decision-making and rapid response from users.
One primary goal of this PhD project is to tackle these challenges by developing a sophisticated operational model for user intent recognition. This model will be grounded in the analysis of motion and biological signals, such as position, interaction force, and electromyography (EMG) data. By accurately identifying and predicting the tasks users perform and their target locations in real-time, the model aims to enhance the efficiency and fluidity of human-robot interactions.
Building upon this predictive model, the project will also focus on designing an assistive control algorithm for collaborative robots. This algorithm will possess two key characteristics: robustness, to tolerate fluctuations and errors in user operations, ensuring smooth task execution; and adaptability, to automatically adjust assistance strength based on the robot's certainty in recognizing user needs as well as other factors such as users’ attention. The candidate will carry out user studies in human-robot collaborative tasks such as disassembling to evaluate the effectiveness of the developed technologies.
The applications of this project are broad and varied, encompassing scenarios like remote operations in nuclear decommissioning, surgical assistance, and rehabilitation robotics. The overarching aim is to establish a more user-centric approach to human-robot interaction, one that is fundamentally based on a nuanced recognition of user needs and intentions.
Location and Facility:
The successful candidate will work at the RAICo One facility in Cumbria, where they will have the opportunity to work in close partnership with robotics, control and AI experts from the University of Manchester, Sellafield Ltd and the UK Atomic Energy Authority.
The facility is well equipped with a range of robotic systems, including quadruped robots like Unitree B1 and Boston Dynamics Spot, robotic manipulators such as Kuka LBR iiwa and Kinova Gen3, various mobile and aquatic vehicles including AgileX Scout, Husarion ROSbot, as well as haptic interface Haption Virtuose6D. RAICo One has also dedicated robotic test areas that include unique demonstration areas for aerial, ground and aquatic robots. The full list of equipment is available on our website https://hotrobotics.co.uk/facilities/university-of-manchester-2/.
Applicants should have, or expect to achieve, at least a 2.1 honours degree or a master’s (or international equivalent) in a relevant science or engineering related discipline.
Experience in user intention recognition, robotic manipulator control, neuromuscular control models, and human-robot interaction will be an advantage. Desirable skills include expert programming (C++ and Python), statistical analysis, experience with ROS and Virtual/Augmented Reality
We recommend that you contact the supervisors before you apply: Dr. Xiaoxiao Cheng (xiaoxiao.cheng@manchester.ac.uk) and Prof. Guido Herrmann (guido.herrmann@manchester.ac.uk).
Type / Role:
Subject Area(s):
Location(s):