By Sean Fenske, Editor-in-Chief

Developing a robotic joint from scratch requires significant knowledge of not just the complete system, but also of its necessary components—encoder, drive, gearbox, sensor, brake, etc. The developer also needs to know how best to configure these parts with each other and understand what settings and programming should be used to best optimize performance.

In addition, precision is paramount. Since the surgical robot will need to position itself in the same location repeatably and accurately, the reliable performance of a joint is critical. Robotic systems are functioning inside of human beings, leveraged for delicate surgical procedures, so they cannot miss a target due to an unreliable joint. It’s not an endeavor to be taken lightly, and wise developers call upon the aid of experts who do this often and are intimately familiar with all the components.

Fortunately, one such expert spoke on this topic in the following Q&A. Peter van Beek, Medical Business Development Manager at maxon precision motors, took time to expand on just how complex this process can be, what designers should keep in mind when beginning to develop a surgical robotic system, and the benefits of working with a vertically integrated manufacturer of robotic joints.

Sean Fenske: First, can you please explain how medical robotic drive assemblies can be made modular in their design for medical robotics?

Peter van Beek: When maxon refers to a “modular” design approach, it infers using standardized components that are configured a make a customer-specific robotic joint for a certain location and application within a larger robotic system. Robotic drive assemblies are comprised of a DC brushless motor (engine), strain wave or planetary gearbox, optical and magnetic-based sensors, brakes, output bearing assembly, and drive electronic components.

Beyond the customer’s component selection for a particular axis, the gear ratio and sensor resolution must be determined, and whether torque sensing at output or a brake is required for a particular axis. A manufacturer like maxon, which is vertically integrated and engineers all the sub-components, is familiar with how to seamlessly combine these parts as well as validate and qualify all the interfaces within the joint. Concurrently, the housing needs to be developed to ensure protection and thermal management (e.g., heat sink), while also making the final solution as compact and power dense as possible.

Fenske: Does modularity enable a robotic joint to be used for different medical applications and surgical robotic systems that are otherwise different?

van Beek: Yes, the same robotic joint can be used for different medical applications, but typically they are customized to some degree for each axis (cabling, mounting, and selected components). In some cases, the same joint can be used in different areas of the robot, which allows economies of scale and a lower overall cost per system. The modular approach results in a complete robotic joint assembly that can be simply installed, thus allowing the roboticist to fully concentrate on core robotic tasks and challenges instead of combining components from different manufacturers.

Early in development, it may make sense to assemble small quantities of your robotic joints (i.e., functional samples), but I suggest leaving larger volume production to a trusted supplier of robotic joints. Place the responsibility of assembly of all sub-components, critical interfaces, documentation, quality validation, and future obsolescence issues with maxon. This is the best method to ensure consistent quality and reduce headaches with the larger design. You build the robot, maxon will provide the best-in-class engines.

If all the components are produced by the same vertically integrated manufacturer, the supplier is designing the various individual components with modularity (combinability) and the final robotic joint in mind. This is a more favorable alternative to a company combining components from several different manufacturers that are not necessarily designed to be compatible with other components. In other words, if you’re not relying on a vertically integrated manufacturer to work with, you’re attempting to design a robotic joint from scratch and trying to make all the components play nicely together—not an easy task.

Fenske: What attributes is a roboticist or surgical robotic designer seeking in a robotic joint?

van Beek: This is an excellent question. Regardless of which components are selected and the final form of a specific joint, quality is always paramount and critical for medical robotics. The complete robotic system, which utilizes robotic drives, performs critical tasks in and around sensitive parts of the body, such as nerves, vasculature, and organs.

One of the more demanding applications using drive assemblies is surgery procedures involving the eye. The robotic joints must perform flawlessly during the procedure when functionality, positional accuracy, and repeatability are essential. Medical robotics commonly utilize strain wave gears, which are compact, lightweight, have zero backlash, and can transfer high torque. These types of gears provide the positional accuracy, repeatably, and torsional rigidity required.

Finally, to answer your question directly, the following attributes are desired: reliability, efficiency, compactness, light weight, extreme power density, audibly quiet, dependable, and long-lasting.

Fenske: What are the major challenges facing a manufacturer that provides modular customization of robotic joints to a large customer base?

van Beek: Three that immediately come to mind are manufacturing sub-components that allow seamless integration with mating components, achieving consistent high quality, and producing a final robotic joint capable of lifetimes surgical robotic maker’s demand. The modular approach only works well with great assembly processes, well-trained assemblers, fixtures, tooling, and complete end-of-line testing.

In addition, one of the most significant challenges of building customized robotic drive assemblies is handling customer-specific cabling and connector sub-assemblies, used to connect power and communication from the joint to the drive electronics. Often, this custom sub-assembly is a major determinant of lead time for the entire joint.

Another challenge is stocking the many different diameter brushless motors (each having two length variants) and different gear ratios (for each diameter). There is a substantial amount of internal structure setup that summarizes all the base components used in a particular build, in addition to any add-on components like brakes or torque sensors. This setup includes the customer-selected resolution of the encoders, which are programmed during production.

Fenske: Do you have any additional comments you’d like to share based on any of the topics we discussed or something you’d like to tell medical device manufacturers?

van Beek: When deciding whether to construct your own robotic joint or to simply purchase one, it is important to consider the long-term or well beyond proof-of-concept samples. The time between the start of a project (i.e., initial functional samples) to serial production (i.e., large volumes of joints) can take many design cycles and years to complete. Moving your internally designed robotic joints from proof-of-concept to governing body submission will require a dedicated project engineer to handle the customization of components to allow coupling together, while also focusing on obsolescence, validation, and multiple vendor relationships. It is complicated. Leave these tasks to maxon, which has been doing such integrations for more than 60 years.

Click here to learn more about maxon precision motors >>>>>