Development, modelling and control of micro-scale  steerable soft growing robots for intraluminal interventions 

Project reference:  SIE_01_22
First supervisor:  Christos Bergeles
Second supervisor:  Kawal Rhode

Start date: October 2022

Project summary:  Soft growing robots can elongate from their tip through a process of inversion, while also conforming to their environment due to their inherent softness. Soft growing robots are a recent innovation but have already showcased significant potential in the navigation of sensitive environments, such as archaeological sites and the human body. 


This project aims to push the boundaries of catheter-like soft robotics through innovation in their steerability, manufacturing and sensorisation. The student and research team will develop manufacturing methods for long micro-scale soft growing robots that can act as intraluminal catheters, with an eye on intravascular, intracerebral, and intraspinal applications. In addition, research will be conducted on the steering ability of such robots, be that through micropatterned soft internal catheters, pneumatic chambers, or tendons. Finally, the student will investigate sensorisation of the soft robot body, exploring tomographic algorithms, and optical sensing methods. 

Project description:

Robotics and imaging will continue to revolutionize healthcare by overcoming limitations of existing medical practice and enabling the clinical translation of emerging treatments. The project will develop and demonstrate for the first time a minimally-invasive robotic platform that will provide clinically innovative access to the Central Nervous System (CNS) via the Ventricular System (VS). Specifically, a miniature soft steerable robot, driven by biomimetic apical extension methods, will navigate safely through the spinal subarachnoid space under fluoroscopic guidance and reach target locations in brain ventricles, where functions such as payload delivery, biopsy, or cauterisation can be delivered. 


In our proposed robotically assisted interventional procedure the clinician will insert a soft-growing robot in the lumbar region of the spine. Under fluoroscopic guidance and by employing apical extension and steering, the clinician will guide robot extension through the subarachnoid space while complying to the surrounding structures (nerve roots, ligaments, arteries). The robot will enter the fourth brain ventricle via the Magendie aperture and subsequently extend to its target location anywhere in the brain ventricles. Once there, we envision the robot to locally deliver drugs, as required to maximise the therapeutic potential of pharmaceuticals aiming to treat, for example, Alzheimer. 


The project has three primary objectives: 

Creation of manufacturing methodologies for sub-millimetre soft steerable growing robots: Our clinical application requires robots of 0.5mm diameters, with extension capabilities of up to 2m. Since conventional soft growing robot manufacturing approaches are not suitable for this, the PhD student will investigate bespoke manufacturing methodologies based on localised LDPE sheet approximation.  

Sensorisation of soft growing robot structure: Robot growth is pressure driven, where pressure causes the eversion of new material from the robot tip. While the robot is soft and can comply to the anatomy, excessive pressure may cause damage to nerves and ligaments. The project will investigate smart materials that endow the robot’s structure with tomographic capabilities, force sensing, and/or promote precision in image-based localisation. 


Model-based robot control: The final step of the project will be modelling and control of the developed robotic system within anatomical phantoms. Modelling will be based on our groups Reduced Order Modelling approaches and open-source package TMTDyn, while control will make use of those models in a Model Predictive Control framework.