About robots, robot development
and those who make it happen


Dexterous Manipulators in Hazardous Environments

May 06, 2015

Oxford Technologies Dounreay Shaft Intervention Platford
Oxford Technologies Dounreay Shaft Intervention Platford Picture: Oxford Technologies

Dexterous Manipulators in Hazardous Environments – Presentation to ERF by Oxford Technologies

At the recent European Robotics Forum (ERF) in Vienna, Stephen Sanders, Director of remote handling consultancy, Oxford Technologies, gave a presentation on the use of dexterous manipulators in hazardous environments, based on his company’s extensive experience in this field.

The first European Robotics Forum was held in San Sebastian in 2010.  Since that time the conference has become a prestigious annual event, attracting over 300 scientists, engineers, academics and robotics professionals from the European Commission and providing an opportunity for specialists and researchers to gather the latest information and strengthen the potential of European robotics.

Stephen Sanders’ presentation on the use of dexterous manipulators in hazardous environments looked at the topic from the initial definition of such environments, through to current examples of remote manipulation projects. 

An increasing number of scientific, manufacturing and defence applications are creating the need for specialized remote handling solutions to ensure the health and safety of the operatives and engineers who operate and maintain technical components in hazardous situations such as areas of high radiation, space, sub-sea and many other toxic environments.  

One example discussed at length was the remote manipulation and maintenance of fuel handling components in liquid metal (Lead-Bismuth) in the MYRRHA Hall.  MYRRHA is a multi-purpose flexible irradiation facility, designed to replace the BR2 materials testing reactor, and is one of the corner stones of the European Research Area of Experimental Reactors.  Due to the highly hazardous environment, maintenance operations on the primary machine systems and equipment are to be preformed using remote handling. All services entering the MYRRHA machine, including the beamline, will require remote maintenance. The recommended solution includes the implementation of two bi-lateral servo-manipulators, working in a master-slave mode. The slave manipulators will be remotely operated using kinematically identical master manipulators supported with closed cycled television feedback. 

Other examples mentioned, included irradiated waste removal using remote manipulation down a 65m deep shaft (in water) at Dounreay, metal cutting underwater, in a nuclear pond at Sellafield, and maintenance work in the Divertor and Hot Cell at the ITER Nuclear fusion Tokamak.

Stephen then went on to look at the overall requirements of remote tasks, dividing this into two parts, the first being the need to design new plant to accommodate remote maintenance – in-line with a Remote Handling “Code of Practice”, from the beginning, which can therefore be maintained by “baseline” RH equipment.  The second part being the need to create solutions for situations where the need for remote handling had not been designed-in, and which therefore require the development of special (non-baseline) RH equipment or tools.

Remote manipulators and their tooling must be designed to meet various criteria, the first being the ability to survive in the specific hostile environment in which they will operate, which requires the selection of appropriate materials and components, as well as the ability to meet a wide range of remote task requirements.  Above all it is necessary to ensure that the manipulators are reliable, easy to maintain and easy to decontaminate.

Stephen suggested three key requirements for a successful RH approach: firstly the need for simplicity and adaptability, taking as a starting point the consideration of how the same task would be done in the absence of a hostile environment and keeping the RH system as technologically simple as possible.   Secondly, the need to bear in mind the need to recover the RH equipment in the event of failure, and to make allowances for the possibility that the plant may not be in the same condition as when it was installed, and thirdly, to guard against the assumption that every design solution should be entirely bespoke and/or involve a robot.

Recently Oxford Technologies have been trialing stereoscopic vision systems, with the objective of investigating the impact of using a 3D stereoscopic camera system, in tandem with a traditional 2D camera view, to aid Dexter™ operators when performing remote handling tasks. Dexter is the company’s advanced servo-manipulator master-slave system, which has been successfully used for highly dexterous remote maintenance tasks in nuclear fusion research and which is available in single or twin arm configurations.

The stereoscopic vision system trials were conducted with cameras that mimic the image definition and output signal format of a typical radiation tolerant camera that would be used within the ITER environment.  Additionally the trials were conducted with high definition cameras in anticipation of radiation tolerant cameras eventually being able to achieve this level of resolution. 

A series of remote handling tasks were undertaken within a mock-up environment and these were timed to assess operator performance when performing the tasks in both 2D and 3D mode.  Tasks included the assembly of small loose electro-mechanical parts, fixing a tie-wrap, cleaning/paining with a brush and handling discrete electrical components.

Six people with varying levels of Dexter experience were used to undertake the trials, some trialists performed the tasks in 2D mode first and then 3D mode, whilst others worked in the opposite order, starting in 3D mode and then performing the same tasks in 2D mode.   The trials are currently ongoing, it is intended that the results will be compared against a traditional 2D camera system.

Concluding his presentation Stephen stressed that the success of an Remote Handling project relied on three things, the importance of developing an adaptable RH approach, the need to bear in mind that the RH system will need to be remotely recovered in the event of its failure, and the necessity of expecting the unexpected in these very hazardous environments. 

Source: Oxford Technologies
Sign in

X Close Panel
Forgot password?