Are you applying to universities? | SHARE YOUR EXPERIENCE Are you applying to universities? | SHARE YOUR EXPERIENCE

Combining Learning-based and Sampling-based Approaches for Manipulation Planning in Unstructured and Cluttered Settings [Self-Funded Students Only]

   Cardiff School of Computer Science & Informatics

This project is no longer listed on and may not be available.

Click here to search for PhD studentship opportunities
  Dr Juan Hernandez Vega, Dr Z Ji  Applications accepted all year round  Self-Funded PhD Students Only

About the Project

Grasping and manipulating objects are some of the basic and most common tasks that a robot (such as manipulator arms or mobile manipulators) must do when collaborating with human partners. While it is considered a simple task for humans, sometimes it could be a challenging robot operation. For a robot, grasping an object normally involves perception to detect and determine the grasp affordances of the object [1], as well as planning the motion to approach and grasp the object. With the increasing popularity of machine learning techniques, new frameworks/libraries have been proposed to both automatically detect the best grasping affordances and plan motions to reach the object.

While such learning-based approaches have considerably improved the response time to determine the grasp affordances, they might still fail to plan feasible motions in situations with characteristics that were not fully represented within the training sets.

To cope with this limitation, we propose to combine and benefit from both model-based planning (e.g., sampling-based) and learning-based approaches. To do so, a learning-based framework can be used to detect the best object grasp affordances, while a sampling-based planner that attempts to reach the object can further refine and learn which affordances are feasible according to the robot motion capabilities. As a more specific example, a computer vision pipeline can be used to detect and learn the best grasping poses of an object that might be surrounded by obstacles in cluttered settings. The output of the pipeline is one or more grasping regions around the object or, alternatively, a set of grasping poses. Such grasping regions or poses will be tested by an off-the-shelf motion planner [2], which will inform the grasping feasibility, i.e., whether the provided grasping regions or poses is doable for a robot.

The project will need to research about existing works in grasping poses detection, and how they can be effectively used in conjunction with the robot motion planner. We aim to contribute a new hybrid manipulation approach that combines novel learning-based techniques and sampling-based motion planners. The results of this project are expected to be validated with KUKA LBR iiwa arms or a mobile manipulator system, both available at Cardiff University. This project will also benefit from existing international collaborations with other well-known research institutions (e.g., Rice University in Houston, TX).

Please contact Dr Juan D. Hernandez Vega for further information:

Academic Criteria

A 2:1 Honours undergraduate degree or a master's degree, in computing or a related subject. Applicants with appropriate professional experience are also considered. Degree-level mathematics (or equivalent) is required for research in some project areas. 

Applicants for whom English is not their first language must demonstrate proficiency by obtaining an IELTS score of at least 6.5 overall, with a minimum of 6.0 in each skills component

How to apply

Please contact the supervisors of the project prior to submitting your application to discuss and develop an individual research proposal that builds on the information provided in this advert. Once you have developed the proposal with support from the supervisors, please submit your application following the instructions provided below

This project is accepting applications all year round, for self-funded candidates via 

In order to be considered candidates must submit the following information: 

  • Supporting statement 
  • CV 
  • In the ‘Research Proposal’ section of the application enter the name of the project you are applying to and upload your Individual research proposal, as mentioned above in BOLD
  • Qualification certificates and Transcripts
  • Proof of Funding. For example, a letter of intent from your sponsor or confirmation of self-funded status (In the funding field of your application, insert Self-Funded)
  • References x 2 
  • Proof of English language (if applicable)

Interview - If the application meets the entrance requirements, you will be invited to an interview.  

If you have any additional questions or need more information, please contact:  

Funding Notes

This project is offered for self-funded students only, or those with their own sponsorship or scholarship award.
Please note that a PhD Scholarship may also available for this PhD project. If you are interested in applying for a PhD Scholarship, please search FindAPhD for this specific project title, supervisor or School within its Scholarships category.


[1] Ardón, P.; Pairet È.; Petrick, R.; Ramamoorthy, S.; and Lohan, K. S. "Self-Assessment of Grasp Affordance Transfer". IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). October 2020.
[2] Hernández, J. D.; Moll, M.; and Kavraki, L. E. “Lazy evaluation of goal specifications guided by motion planning,” in IEEE International Conference on Robotics and Automation (ICRA), 2019, pp. 944–950.

How good is research at Cardiff University in Computer Science and Informatics?

Research output data provided by the Research Excellence Framework (REF)

Click here to see the results for all UK universities
Search Suggestions
Search suggestions

Based on your current searches we recommend the following search filters.

PhD saved successfully
View saved PhDs