Preceyes set ambitious objectives for its surgical system. Not only would the enhanced precision make new procedures possible, it would also improve existing procedures, such as the peeling of membranes from the retina and the replacement of the eye’s fluid. In working toward these objectives, the Preceyes team had two priorities: to ensure patient safety and to produce a working clinical investigation device as efficiently as possible.

To achieve their design objectives within the development schedule and safety constraints, Preceyes engineers needed to take advantage of modeling and simulation technology that had been used in other safety-critical applications across multiple industries. Specifically, they needed to model complex control logic and then perform functional verification of their design, first via simulation and then via real-time tests on the actual robot. In addition to real-time control software, the team needed to develop a non-real-time application with an interface running on a PC that physicians could use to configure settings before and during surgical procedures. 

Preceyes used Model-Based Design with MATLAB, Simulink, and Simulink Real-Time to accelerate the development of a real-time control system for its PRECEYES Surgical System.

While constructing a basic model of the robot’s kinematics and dynamics, Preceyes engineers conducted system identification tests in which they used Simulink Real-Time to produce excitation signals for the robot’s 11 electromotors and then measured the responses across each of its 11 degrees of freedom.

After analyzing and fitting the measurement data in MATLAB, they used the results to develop a Simulink plant model that described how the robot moved in response to applied forces.

The team developed a controller model that processes input from the system’s joystick and approximately 60 sensors—including optical and magnetic encoders—and generates the necessary motor signals to move the tool tip in response to the surgeon’s movements. Within this model, they incorporated EtherCAT blocks from Simulink Real-Time to interface with nodes on the robot’s EtherCAT network.

Using Stateflow^®, the team modeled sequential logic for the system’s various operating modes, which include calibration, initialization, and self-verification.

After verifying the basic operation of the controller and control logic through simulation, the engineers generated code from their model using Simulink Coder™ and deployed the compiled code to a real-time target computer running Simulink Real-Time. This computer was connected to the robot’s sensors and motors via the EtherCAT network.

To refine and enhance the controller, the team made improvements to the model, checked them via simulation, and then tested them on the actual robot using the Simulink Real-Time setup.

The engineers used GUIDE, a MATLAB UI development tool, to develop the touch-screen application software that surgeons use to change surgical settings and receive visual and acoustic feedback during surgery.

The safety and efficacy of the PRECEYES Surgical System has been demonstrated in 14 surgical procedures. Additional demonstrations are planned at top surgical sites, including a two-year collaboration with Rotterdam Eye Hospital. The company is also working on CE registration and the next generation for volume production.

• Core controller developed by one engineer. “With MATLAB and Simulink, I did not have to program a low-level architecture for the controller by myself. As the sole software engineer developing the first release, that was a huge advantage—in fact, I doubt if a single engineer could have done the work otherwise,” says Beelen. “The software was extensively reviewed by consultants, which was quite easy because of the readability, requirement traceability, and report generation capabilities in Simulink products.”
• Patient safety assured. “I always said I would not bring a device to market unless I was fully confident using it on myself or on a loved one,” notes Beelen. “With the safety mechanisms we had implemented in Simulink, I was 100% comfortable that there would be no problems with the software.”
• Road map to industrialization set. “For the development of the industrialized release of our system, we will save time by generating code from our existing controller model, using Embedded Coder to target an embedded processor,” says Beelen. “We will follow a more formal development and testing workflow, with improved revision control and a larger development team, using model-based testing, verification, and validation capabilities in Simulink.”