Tag Archives: innovation

Connecting robotic touch and vision

Leave a reply

katherine kuchenbeckerSome months ago I wrote about soft robots that could delicately pick up fragile objects [see my post entitled ‘Robots with a delicate touch’ on June 3rd, 2015]. These robots, developed by George Whiteside’s research group, went some way towards mimicking the function of our hands.  However, these robots are numb because they have no sense of touch.  Think about how hard it would be to strike a match or pick up an egg without your sense of touch. Katherine Kuchenbecker from the University of Pennsylvania is working on robots with tactile sensors that detect pressure and vibrations.  This sensitivity transforms their ability to perform delicate tasks such as picking up an egg, or perhaps more significantly perform surgery.  I listened to Professor Kuchenberger speak at a meeting at the Royal Society on ‘Robotics and Autonomous Systems’ where she put us off our lunch with some gory videos on robot-assisted surgery. You can watch them at her website. Her vision is of robots that connect vision and touch, which is of course what we do effortlessly most of the time.

Insidious damage

1 Reply

bikeRecently, my son bought a carbon-fibre framed bike for his commute to work. He talked to me about it before he made the decision to go ahead because he was worried about the susceptibility of carbon-fibre to impact damage. The aircraft industry worries about barely visible impact damage (BVID) because while the damage might be barely visible on the accessible face that received the impact, within the carbon-fibre component there can be substantial life-shortening damage. I reassured my son that it is unlikely a road bike would receive impacts of sufficient energy to induce life-shortening damage, at least in ordinary use. However, such impacts are not unusual in aircraft structures which means that they have to be inspected for hidden, insidious damage. The most common method of inspection is based on ultrasound that is reflected preferentially by the damaged areas so that the shape and extent of damage can be mapped. It is difficult to predict the effect on the structural performance of the component from this morphology information so that, when damage is found, the component is usually repaired or replaced immediately. In my research group we have been exploring the use of strain measurements to locate and assess damage by comparing the strain distributions in as-manufactured and in-service components. We can measure the strain fields in components using a number of techniques including digital image correlation (see my post entitled ‘256 shades of grey’) and thermoelastic stress analysis (see my post entitled ‘Counting photons to measure stress‘). The comparison is performed using feature vectors that represent the strain fields, see my post of a few weeks ago entitled ‘Recognising strain’. The guiding principle is that if damage is present but does not change the strain field then the structural performance of the component is unchanged; however when the strain field is changed then it is easier to predict remanent life from strain data than from morphology data. We have demonstrated that these new concepts work in glass-fibre reinforced laminates and are in the process of reproducing the results in carbon-fibre composites.

Sources

Patterson, E.A., Feligiotti, M., Hack, E., 2013, On the integration of validation, quality assurance and non-destructive evaluation, J. Strain Analysis, 48(1):48-59.

Patki, A.S., Patterson, E.A., 2012, Damage assessment of fibre reinforced composites using shape descriptors, J. Strain Analysis, 47(4):244-253.

Greed overwhelms ethics

3 Replies
My car but not my house!

My car but not my house!

The scandal about Volkswagen emissions has already caused journalists and others to wring their hands or to preach sermons, or both, about the ethical standards of the engineering profession, see for example the Editor’s blog in Professional Engineer where he reminds us that professional engineers should conduct their professional work and relationships with “integrity and objectivity and with due regard for the welfare of the people, the organisations and the environment with which they interact”. The quotation is from the Royal Charter of the Institution of Mechanical Engineers. Fine words written to induce good intentions that rarely survive in the face of greed or simply the need to keep your job so that you can feed and house your family.

In my view, the emissions scandal seems to have parallels with the banking scandals of the past decade, in which corporate greed has trampled over ethics and morals in the pursuit of ever larger profits while government regulators through incompetence or acquiescence have allowed it to continue. Volkswagen were wrong to design a device to cheat the Government emission tests, but Government regulators were naive to design a test in which it was so easy to cheat. Senior executives at Volkswagen have blamed their employees following other recent examples.  These company leaders were paid gigantic salaries to provide both leadership and management and, in my opinion, they have failed in both by blaming the people they are supposed to be leading and by allowing the scandals to happen in the first place.

The evidence would suggest that we can trust neither corporations nor governments to take care of the environment. One solution is for all vehicles to provide real-time information on the dashboard about NOX and Carbon emissions as well fuel consumption then we can make our own choices. When I moved to the US more than a decade ago, I was surprised to find that fuel consumption data was not available on the dashboard of most cars as it was already commonplace in Europe. Perhaps its presence was a factor in the development of fuel efficient cars in Europe although clearly higher fuel prices play a large role.  However, in the absence of a tax on emissions, real-time emission data on the dashboard would motivate engineering ingenuity to compete to produce lower emission designs instead of wasting creativity on cheating in useless Government tests.

Here are some of facts to support my statements above about large profits and gigantic salaries:

Volkswagen profits rose 21% in 2014 to more than $12billion on an annual turnover of $230 billion which is comparable to the GDP of Portugal.

The CEO of Volkswagen was paid almost $18million per year which is about 250 times the average salary of a Volkswagen engineer.

Sources:

Hibbert, L., Editor’s comment: October 2015 The Volkswagen emissions testing scandal has put the issue of professional ethics in the spotlight, Professional Engineer, October 07, 2015.

Wall Street Journal, Feb 27th, 2015

Volkswagen AG Annual Report 2014

World Bank Data Bank

Reuters, March 12th, 2012

glassdoor.com

 

Death knell for the lecture?

2 Replies
Author lecturing in Yonsei University, Korea

Author lecturing in Yonsei University, Korea

This week I have started filming short video clips for a MOOC that will be broadcast in February in parallel with my undergraduate course on Thermodynamics. The Massive Online Open Course (MOOC) is provisional titled: ‘Energy – Using it and Losing it: Real-World Thermodynamics for Beginners’ and will be offered through FutureLearn to a worldwide audience. The video clips, which essentially replace the traditional 50-minute lecture, will be about 3 minutes long recognising that this is the longest time period that many young people will focus uninterrupted on a single activity.

Last week was the start of a new academic year in which we have been instructed to use newly-installed software and hardware to record or, in the new terminology, video-stream all of our lectures. The ‘streamed’ lectures will be made available online for students to watch at anytime during the academic year. All of this is happening when attendance at lectures is falling, which leads me to wonder whether these events represent the death knell of the traditional university lecture?

We have known for sometime that people’s maximum attention span was typically fifteen to twenty minutes and yet lectures have remained stubbornly at 50 minutes duration with many double lectures timetabled. Considerable ingenuity, imagination and energy is needed to deliver lectures that engage students for these time periods (see Engage Engineering for tips on how to do this). So it should come as no surprise that many lectures are half empty when students have alternatives such as short video clips available online, streamed lectures that can be fast-forwarded over the boring bits or rewound to repeat important sections, as well as the old-fashioned approach of reading a good textbook and teaching yourself.

Lectures are in many ways a theatrical performance, though factual rather fictional. Theatre has had to evolve and adapt in order to survive the advent of cinema, television and most recently the internet. In the process, some theatres and drama companies have disappeared. I think the same is likely to happen with the university lecture – some will evolve and adapt, for instance by embracing new technology, but others will disappear as students choose more effective means of acquiring knowledge and understanding.