CorteX

Future-Proof Interoperable Robotics

Within robotics for nuclear applications, there is a need for a software platform that can adapt to new challenges, new user requirements, and new advances in technology. In order to meet these requirements, RACE is working towards providing a future-proof communications and control platform, essential for future large scale nuclear robotic facilities.

CorteX in action

RACE draws upon the experience gained from 25 years and over 40,000 hours spent maintaining the Joint European Torus (JET) experiment. Over this time, hardware has become obsolete, maintenance requirements have changed, operations have become more complex, and the original Remote Handling equipment has struggled to keep up. Both the hardware and the software have had to adapt to these changing requirements, but at the cost of reduced RAMI (Reliability, Availability, Maintainability, Inspectability).


Nuclear robotics in complex facilities are comprised of hundreds of interoperating systems. They will become increasingly commonplace as safety and productivity requirements drive facility design and operation. In order to be efficient and operate for decades without significant downtime, future systems will be dependent on automation and the sharing of information.


Within current nuclear robotic applications, there are limitations regarding dependability, reliability, security, and compatibility with regulatory requirements.

Existing network and cloud robotics suffer from various issues such as interoperability, heterogeneity, time-varying network latency, security, multi-robot management, common infrastructure design, Quality-of-Service (QoS), and standardisation.


CorteX is built around a self-describing protocol which contains both the data in the system and supporting meta-data. The structure of data within the system is similar to that found in other robot control system platforms, such as the Robotic Operating System (ROS). However, in RACE’s interoperable solution, this data is complemented with a combined architecture and inheritance type structure, allowing for interpretation and discovery of previously unknown systems.


The project aims to deliver interoperable communication standards and libraries, control methods applicable to current robotic technologies, and validation routines to test the stability of the developed platform.

The TARM in the RACE work hall

CorteX's design principles

CorteX, RACE’s in-house developed robotics control platform, has been focused around five key design principles, namely reusability, extensibility, modularity, standardisation, and integrated user interfaces.


  • To guarantee reusability, control solutions must be implemented in a generic fashion in order to be agnostic of their application.
  • These control solutions must also be implemented in such a way to allow them to be extended by later applications, which may require additional functionality.
  • Modularity ensures the ability to replace components of a control system with newer or alternative counterparts, providing the ability to adapt to changing requirements.
  • Interoperability is essential for large-scale integration applications and can only be achieved using agreed standards. However, these standards should not restrict the capability of a platform and therefore need to be customisable while maintaining backwards compatibility.
  • If these four principles are followed, it is then possible to create standard integrated user interfaces to all systems, providing users and developers with a standardised experience.


CorteX

Future-Proof Interoperable Robotics

Within robotics for nuclear applications, there is a need for a software platform that can adapt to new challenges, new user requirements, and new advances in technology. In order to meet these requirements, RACE is working towards providing a future-proof communications and control platform, essential for future large scale nuclear robotic facilities.

CorteX in action

RACE draws upon the experience gained from 25 years and over 40,000 hours spent maintaining the Joint European Torus (JET) experiment. Over this time, hardware has become obsolete, maintenance requirements have changed, operations have become more complex, and the original Remote Handling equipment has struggled to keep up. Both the hardware and the software have had to adapt to these changing requirements, but at the cost of reduced RAMI (Reliability, Availability, Maintainability, Inspectability).


Nuclear robotics in complex facilities are comprised of hundreds of interoperating systems. They will become increasingly commonplace as safety and productivity requirements drive facility design and operation. In order to be efficient and operate for decades without significant downtime, future systems will be dependent on automation and the sharing of information.


Within current nuclear robotic applications, there are limitations regarding dependability, reliability, security, and compatibility with regulatory requirements. Existing network and cloud robotics suffer from various issues such as interoperability, heterogeneity, time-varying network latency, security, multi-robot management, common infrastructure design, Quality-of-Service (QoS), and standardisation.


CorteX is built around a self-describing protocol which contains both the data in the system and supporting meta-data. The structure of data within the system is similar to that found in other robot control system platforms, such as the Robotic Operating System (ROS). However, in RACE’s interoperable solution, this data is complemented with a combined architecture and inheritance type structure, allowing for interpretation and discovery of previously unknown systems.

The TARM in the RACE work hall

The project aims to deliver interoperable communication standards and libraries, control methods applicable to current robotic technologies, and validation routines to test the stability of the developed platform.

CorteX's design principles

CorteX, RACE’s in-house developed robotics control platform, has been focused around five key design principles, namely reusability, extensibility, modularity, standardisation, and integrated user interfaces.

  • To guarantee reusability, control solutions must be implemented in a generic fashion in order to be agnostic of their application.
  • These control solutions must also be implemented in such a way to allow them to be extended by later applications, which may require additional functionality.
  • Modularity ensures the ability to replace components of a control system with newer or alternative counterparts, providing the ability to adapt to changing requirements.
  • Interoperability is essential for large-scale integration applications and can only be achieved using agreed standards. However, these standards should not restrict the capability of a platform and therefore need to be customisable while maintaining backwards compatibility.
  • If these four principles are followed, it is then possible to create standard integrated user interfaces to all systems, providing users and developers with a standardised experience.