Interventional bronchoscopy is an important approach for the diagnosis and treatment of lung diseases. In order to further improve its efficiency and applicability, a robotic platform integrated with novel designs on both hardware and software is presented in this thesis.

A continuum manipulator with interlocking joints is first introduced, which tries to maximise the flexibility and minimise wall thickness. It could serve as an extension of a normal bronchoscope, and guide instruments like biopsy forceps to small airways in peripheral regions.

Next, a robotic bronchoscopy system is developed. It features infinite rotation and continuous translation of the insertion tube, making the operation more convenient and less tiring. This is achieved by the cooperation of a motorised tube feeder with tilted friction wheels and an actuation handle with differential drive mechanism.

Fully automatic navigation with the above robotic bronchoscope is then studied. By combining model-less control framework and model-based Jacobian estimation, the proposed controller is capable of being adaptive as well as robust while steering the scope inside the complex bronchial tree.

Last but not least, two shape sensing approaches for bronchoscopic instruments are also investigated. The first one uses endoscopic images and an electromagnetic sensor to estimate the pose in 3D space. The second method further realises simultaneous shape sensing and tip force estimation by building an accurate mathematical model that considers instrument elasticity, driven cable interactions, and tip force effects.

The researches presented in this thesis are promising to help build the next generation bronchoscopic instruments with enhanced reachability to peripheral airways, improved operation experience, advanced control assistance, and smart sensing of position, shape and force.