November 2019

The opening keynote speaker at the 2019 Condition Monitoring and Diagnostic Engineering Conference (COMADEM) warned against recent claims that the future of predictive maintenance should be led by data science, and not engineers. Professor Andrew Ball, a renowned expert in the field of diagnostic engineering, delivered the first of the keynote presentations at the COMADEM Congress, which was attended by over 180 delegates and held at the University of Huddersfield in the UK. Professor Ball, who also co-chaired the conference, told delegates that the separate techniques of detecting, diagnosing, assessing the severity and prognosis of machine faults require engineering expertise and context to achieve the accuracy and timely results demanded in the field of predictive maintenance. “I have attended conferences recently where speakers have talked about purely data-driven approaches to predictive maintenance, with no concept of what engineering really needs,” he told his audience. Professor Ball stressed that successful interventions could only be achieved by engineers and data scientists working together. “Data-driven methods are truly excellent for identifying patterns and anomalies in large, complex data sets, and warn us when to undertake fault diagnosis, location and severity assessments,” he said, “but the latter steps cannot be achieved using data-driven methods alone. Predictive maintenance is an engineering discipline. One that can be significantly assisted by data science, but only if they work together.” COMADEM, now in its 32nd year, made its second visit to the University of Huddersfield, previously coming to the campus in 2012. Designed as a leading international forum for industrialists, engineers and exhibitors, the conference attracted a worldwide audience with papers presented by speakers from as far away as China, Australia and the United States. Under the theme of Digital Enabled Asset Management, conference papers covered all areas of condition monitoring and maintenance, including signal and image processing, pattern recognition, finite modelling and simulation, as well as root cause analysis, sensors and actuators, asset management, and education and training. News 0 6 | Plant & Works Engineering www.pwemag.co.uk November 2019 Need for engineers and data scientists to work together Professor Andrew Ball A new generation of swarming robots which can independently learn and evolve new behaviours in the wild is one step closer, thanks to research from the University of Bristol and the University of the West of England (UWE). The team used artificial evolution to enable the robots to automatically learn swarm behaviours which are understandable to humans. This new advance published today [Friday 23 August] in Advanced Intelligent Systems, could create new robotic possibilities for environmental monitoring, disaster recovery, infrastructure maintenance, logistics and agriculture. Until now, artificial evolution has typically been run on a computer which is external to the swarm, with the best strategy then copied to the robots. However, this approach is limiting as it requires external infrastructure and a laboratory setting. By using a custom-made swarm of robots with high-processing power embedded within the swarm, the Bristol team were able to discover which rules give rise to desired swarm behaviours. This could lead to robotic swarms which are able to continuously and independently adapt in the wild, to meet the environments and tasks at hand. By making the evolved controllers understandable to humans, the controllers can also be queried, explained and improved. Lead author, Simon Jones, from the University of Bristol’s Robotics Lab said: “Human-understandable controllers allow us to analyse and verify automatic designs, to ensure safety for deployment in real- world applications.” Co-led by Dr Sabine Hauert, the engineers took advantage of the recent advances in high-performance mobile computing, to build a swarm of robots inspired by those in nature. Their ‘Teraflop Swarm’ has the ability to run the computationally intensive automatic design process entirely within the swarm, freeing it from the constraint of off-line resources. The swarm reaches a high level of performance within just 15 minutes, much faster than previous embodied evolution methods, and with no reliance on external infrastructure. Dr Hauert, Senior Lecturer in Robotics in the Department of Engineering Mathematics and Bristol Robotics Laboratory (BRL), said: “This is the first step towards robot swarms that automatically discover suitable swarm strategies in the wild. “The next step will be to get these robot swarms out of the lab and demonstrate our proposed approach in real-world applications.” By freeing the swarm of external infrastructure, and by showing that it is possible to analyse, understand and explain the generated controllers, the researchers will move towards the automatic design of swarm controllers in real-world applications. In the future, starting from scratch, a robot swarm could discover a suitable strategy directly in situ, and change the strategy when the swarm task, or environment changes. New generation of swarming robots

RkJQdWJsaXNoZXIy MjQ0NzM=