Copy
View this email in your browser
Welcome to the fourth issue of the Pipebots newsletter.  Since our last newsletter was issued in June 2020, we have seen some lockdown restrictions being lifted, although the laboratory, experimental and manufacturing facilities at the UK universities remained closed.  During the past few months the Pipebots team have continued to work remotely, working from home and meeting via teleconferencing facilities.  We continued to run our series of webinars amongst the team, discussing various topics including localization for robots in pipes, integration, communication and power, autonomous control and robotic simulation.  The webinars have been well attended and enjoyed by the research team, and this is something we are continuing during the next few months. 

We are pleased to report that the laboratory and experimental facilities are starting to reopen and the research teams are slowly getting back on site to resume their experimental work.  A priority for the team is to complete testing of the Sprintbot produced as a result of Sprint 2 which had been delayed due to the team being unable to access ICAIR as a result of the pandemic.

We are very excited to welcome Bryony Bowman to the team!  Bryony joins Theme 7 (Emerging Science) as a PhD Student and will be looking at opportunities for sustainability and resource recovery in the water industry.  Welcome to the team Bryony!

During the last few months, we have also held a number of meetings with stakeholders and we continue to work with and consult our industry partners.  We have submitted a number of joint research proposals with industry partners and the research team have continued to work on publications which we are looking forward to sharing in due course!


We continue to work with Human Studio on exciting an augmented reality and virtual reality project which is a follow up to the Pipebots video.  This will be showcased at Festival of the Mind 17th to 27th September 2020, and will also be made available on our website and YouTube channel.  If you are in Sheffield between 17th to 27th September, pop in to the Millennium Galleries and visit our exhibition:  Could tiny robots end roadworks?
Research Update
Sprintbot: Return of the Sprint! 

The robotics experts over at Theme 3 have been busy this month working on a number of upgrades to sprint-bot; Pipebots first sewage inspection platform.  Sprint-bot was initially the development of a 3-week long hardware sprint, but the sprints conclusion was cut short due to COVID-19.  Now there is hope on the horizon for testing to continue as the ICAIR facility finally starts to re-open. 

In preparation for next month’s testing, Theme 3 will be developing a new tail for sprint-bot.  Sprint-bot is designed like a ball so it requires a tail to stop itself from rolling over, the current tail is too rigid to allow turning in tight spaces. The new proposed tail will be deformable, allowing it to conform to the pipe walls during robot rotation.
We are looking forward to doing some real world testing next month and will be excited to share some results from the hard work that’s been put into Sprint-bot’s development!  Read more HERE
Sensors
Xudong Niu has tested Red pitaya as a sonic prototyping platform for low ultrasonic frequency signal generation and reception with a set of parking sensors at 40 kHz at home. The signal processing using Gaussian chirp signal is working using the platform with a single signal transmission and receiver channel. The algorithm can enhance the signal-to-noise ratio for captured signals effectively in the experimental environment. The further tests are available to use the kit integrated with the ‘Sprintbot’ for evaluating the signal performance in a ‘real’ environment in ICAIR, Sheffield.
Xudong Niu has successfully used the Rayleigh damping in finite element modelling in Abaqus software for the excitation and propagation of a simple guided wave mode to account for ultrasonic wave energy loss with geometric spreading and attenuation.  This model can be used to investigate the scattering of Lamb wave modes on different defect shapes and sizes. The methodology is an effective pathway to reduce the complexity of the analysis of Lamb wave modes propagation in a pipe.

Yicheng has developed an analytical approximation to simulate the sound propagation in a partially filled water pipe. It accounts for the wave dispersion and can be used to study acoustic communication and navigation in pipes This work is about to be submitted for publication in the Journal of Applied Acoustics.
Autonomous Control
Theme 4 has been working on building self-assembly algorithms in a swarm of Pipebots. We are investigating methods of how to assemble Pipebots such as male-female docking mechanism, assigning gripper/grippee or using electromagnet etc.; which shape how Pipebots should be assembled to carry out the assigned tasks; and how can the assembled organism move collectively. We are researching methods to apply and evolve neural networks to address cooperative navigation and morphological computation. 
 
Theme 4 are also working on how to exploit the environment for self-assembly such as using the pipe itself for aligning the Pipebots together for docking (see below) because robot sensors are limited in accuracy to assist efficient docking/connecting Pipebots together.
Theme 4 are using Gazebo and ROS integrated environments to simulate robot interaction and develop the algorithm.
Navigation
 
Theme 5, working on Navigation have had a paper accepted for publication in the 2020 IEEE International Conference on Multisensor Fusion and Integration (IEEE MFI), the paper titled ‘Echo Localization for Pipe Inspection Robots’ was written by Rob Worley, Yicheng Yu, and Sean Anderson.
Communication and Power

Since June Theme 6 has continued to work on the analytical and numerical modelling of electromagnetic wave propagation inside buried sewer pipes. The aim, as before, is to get an understanding and make informed predictions on what distances we can establish communication links inside sewer pipes, using currently available technology.  To achieve this we have looked at the different parameters that affect the propagation of EM waves, namely pipe diameter, nature of surrounding soil, and soil moisture. We have investigated frequencies from 100 MHz to 1 THz in order to get a "big picture" view of what will and what will not be feasible.
 
We have also looked at the possibility of communicating through the soil to base stations and robots on the surface, in order to provide redundancy for the wireless link.  The findings so far indicate a strong preference for higher frequencies, i.e. 6 GHz and above, particularly in the smaller diameter pipes. These will propagate with less losses, meaning individual Pipebots can travel farther from their deployment point. Conversely, using much lower frequencies, on the order of 100 MHz - 200 MHz could provide connectivity even through several metres of soil, depending on its moisture.
 
Going forward, Theme 6 will finalise the modelling work and prepare a full report on that. Field measurements at ICAIR will take place as soon as possible to validate the theoretical and simulation work. These measurements will also be used to make the case for using millimetre-wave frequencies, e.g. 24 GHz and 60 GHz, and corresponding new hardware, for Pipebot-to-Pipebot and Pipebot-to-Base communications.  Theme 6 will also do some prototyping work to test our various ideas, such as dual-frequency communication for within pipe and pipe to ground links, peer-to-peer connections for data relay, and using multiple Pipebots to form synthetic antenna arrays.
Emerging Science
 
Theme 7’s current focus is on the development of system maps that will support this two-way process; and, contribute to the development of a digital-twin for buried pipework assets that can be used as a virtual test bed for robot prototypes.  This is in the context of ‘Sprintbot’, the prototype robot that Pipebots is developing for autonomous surveying of sewer pipes.

The mapping work addresses Theme 7’s principal interfaces, shown below.  Making the business case for change is an essential step in translating research findings into operational improvements.  Understanding existing asset management procedures and the potential impact that any new approach may have on these is crucial to assessing the desirability of any proposed change.  At the same time, economic, political and social drivers/trends must be born in mind, along with the influence they are likely to have on the regulatory environment, extending through to individual company policy.
T7’s principal interfaces
The operational system maps produced will be important components in the formation of the digital twin.  Dŵr Cymru Welsh Water (DCWW) has assisted with modelling by providing access to its operating procedures.  The modelling tool used is CORE9, developed by Vitech Corporation in the United States.  The diagram below shows a typical example of the sort of graphical process description being produced.  SAP MRS is SAP’s multi-resource scheduling tool; MPB refers to its ‘merged planning board’ screen.
Theme 7’s aim is ultimately to create a full suite of system maps that will help to demonstrate objectively to water companies and other asset managers the benefits that Sprintbot can deliver.  Read more HERE
System knowledge generation
Having handed over the Challenge Specification documents to become living documents updated by the whole project team, Theme 8 are moving on to look into how pervasive data will be used.  The first step is focussing on current defect classification metrics used with sewer system CCTV data; and this will also expand into what data is required to feed deterioration models, particularly the useful resolutions for such data (e.g. how accurately should a crack in a pipe be measured).  Condition monitoring of buried pipes inherently difficult, currently the most common form is the use of CCTV in sewers which is expensive and time consuming, so usually is focussed on critical pipes and areas with known problems.  The defect classification systems, developed to interpret the visual data collected by CCTV to describe the condition of a pipe, vary across the world.  Understanding how different countries use defect classification will help to develop methods for handling the large amounts of data from a range of sensors on Pipebots which it is envisaged will eventually trawl entire networks.  Furthermore, researching existing pipe deterioration models can help to focus the data requirements from the pervasive sensing.
Other Research
Robots tackle coronavirus head on!
During lockdown members of the Pipebots team have been putting their skills to good use by creating robots which are designed to tackle coronavirus head on.  The robots combine navigation, computer vision and artificial intelligence (AI) to identify objects that need regular cleaning such as seating, bike stands and door handles.
Once a target object is discovered, it is sprayed with a mist of diluted alcohol. The robots can map the area and are able to manoeuvre without bumping into either static or other mobile objects such as people or animals.   This allows regular cleaning while removing the risk of contamination that a person would face.  The robots have been used in field tests in Leeds city centre, in the lounges at Leeds Bradford Airport and have recently been covered by the BBC and other newspapers.

One of the robots operates on wheels and is capable of covering larger areas – the other has legs for working in more confined or stepped environments.  See the full story HERE.
Upcoming Events
2020 IEEE International Conference on Multisensor Fusion and Integration (IEEE MFI), 14th to 16th September 2020
Festival of the Mind, Millennium Gallery, Sheffield, UK, 17th to 27th September 2020
Get Involved!
The Pipebots team want to engage with a wide range of stakeholders and welcome participation and involvement from academia, industry, governing bodies and investors.  If you would like to participate in one of the case studies, or are interested in part-funding a PhD student, please get in touch:  pipebots-info@sheffield.ac.uk or complete the Pipebots Engagement Form
Twitter
Facebook
Website
Our mailing address is:
pipebots-info@sheffield.ac.uk

Want to change how you receive these emails?
You can update your preferences or unsubscribe from this list.






This email was sent to <<Email Address>>
why did I get this?    unsubscribe from this list    update subscription preferences
Pipebots · University of Sheffield - Broad Lane Building · Broad Lane · Sheffield, Shf S14BT · United Kingdom

Email Marketing Powered by Mailchimp