The main objective of the project is to develop an innovative and universal robotic assistant system to support a human in dextrous manipulation. For this reason we address methods to increase accuracy for lightweight compliant robotic systems during surgical procedures on different levels of autonomy. This varies from telemanipulation to autonomous behaviour. Independent of the control status, a user has the ability to intervene into the treatment process. The approach focuses on adaptive control by exhibiting rich sensory-motor skills and multi-sensory measurement to distinctly increase the system accuracy.
To handle a complex working field a robot system needs a detailed description of its environment. A natural environment is usually fuzzy and continuously changing. Thus it is a major task within this project to find a way to model such environments with sufficient accuracy as far as possible. This has an impact on planning a task since the model might be strongly incomplete. Therefore it is crucial to include a method to describe uncertainties in the planning. An advanced and light-weight robot arm which exploits a redundant structure is one component of the proposed architecture. This allows null-space motion enabling the robot's joints to be reconfigured while the position and orientation of the instrument remains unchanged. Basically, the advanced robot arm includes a complex sensor system and its control devices that are suitable for a wide range of humanoid operations and especially complex surgical procedures.
The system is based on an open and modular architecture. For real operations the configuration can vary from one arm in open surgery to at least three arms in minimally invasive surgery. As the robot is light-weight it can be easily mounted or removed by user during an operation. As highly demanding demonstration activities we will cope with laser osteotomy and palpation.
Goal 1: Increase accuracy of compliant light weight robots through multi-sensory concepts.
Goal 2: Increase accuracy of robotic procedure planning through adaptive models, soft tissue modelling and online motion prediction.
Goal 3: Develop new techniques for laser osteotomy to enable contact-free cutting of arbitrary cutting geometries in the human bone with superior accuracy.
Goal 4: Provide kinaesthetic feedback to robotic systems for palpation in real and simulated environments.