ROBOTIC HANDS: TO WORK ALONGSIDE HUMANS [EN text]

As technology, creativity and business conference LOGIN passed by, we had a great opportunity to talk to Rich Walker, MD of the Shadow Robot Company. Worked over 20 years in robotics, Walker is extremely attached to this industry. He sits on the Innovate UK “Robotics and Autonomous Systems” SIG Advisory Board, which lets him influence the direction the UK takes in robotics in a way that makes sense to SMEs and innovators. Also, he is the Director of EuRobotics, and various EPSRC and University networks and committees around robotics.

The Shadow Robot Company, which Walker runs, is considered as a leader in grasping and manipulation, is constantly developing new robots and applications in the field of cutting-edge robotics and AI. Walker, by himself, says that he likes to find new ways to apply next-generation robotics technology and thinking how to solve real-world problems.

We asked Rich about Shadow Robot Company’s creations and his own view, which way the robots are going.

Can you share with us about your latest projects and what your team is working on now?
There’s a lot going on at the Shadow Robot Company right now including 7 Innovate UK projects and 3 Horizon 2020 projects. Many of our current collaborative projects are around sensing, control, and vision. For example, our EU project COROMA is geared towards the manufacture of metal and composite material parts that can interact both with humans and other machines. We’re also working on a project called iGrasp with Queen Mary University of London (QMUL) to create an autonomous grasping system that can achieve a stable grasp in the most efficient and cost-effective way.

There is continued interest in our robotic hands in farming and agriculture. We previously worked on a project known as AUTOPIC which involved handling ripe strawberries and are now looking forward to working on a similar project.

We’re also excited to be using our hardware within the area of artificial intelligence and have recently partnered with OpenAI so they could develop Deep Learning systems on our hardware, allowing it to
handle objects and learn from failure.

For people, the ideal gripping tool is their five-fingered hand. Should robotics copy the product of nature design or work on designing something entirely different?
Design can play an essential role in how society embraces robotic technology but whether roboticists should copy a product of nature depends on a variety of factors such as the needs and uses of the technology in question. Robots don’t have to resemble humans to be able to perform human-like functions but it certain cases, it may help.

Robots don’t have to resemble humans to be able to perform human-like functions but it certain cases, it may help.

For example, as robotic technology increasingly enters human environments like homes and hospitals, they may be more widely accepted and easier to use if their design is familiar to us. This may not be as necessary with robotics in factories that are needed to perform mechanical tasks.

We’ve achieved commercial success with both starting with the Shadow Dexterous Hand which approximates the kinematics of the human hand – from its overall size to the 20 actuated degrees of freedom along with a further 4 under-actuated movements for a total of 24 joints. The thumb and even the flex of the palm at the base of the little finger resemble that of a human hand with comparable movement range. It can be exploited to generate innovative manipulation control systems or to provide a more defined understanding of the external environment.

We’ve also designed and developed our new Modular Grasper which is more versatile and robust. Its design does not mimic the human hand and yet it is a powerful tool as a comprehensive three-fingered grasper and a cost-effective solution when you need to grasp many different types of objects but only want to use one Hand. The first model has 9 degrees of freedom, position sensing and torque sensing on each joint, ensuring the Hand can make the most accurate and reliable grasp.

In your opinion which profession has the biggest potential to be replaced by robotic hands first?
The correct and mindful use of robotic technology is something we’ve always been conscious of which means using such technology, including our robotic hands, to work alongside humans, as opposed to replacing them and their professions. The idea is to use robotic technology to automate more tasks in the workplace, ones that are repetitively straining, dirty, or unsafe so that it can help free up time for
employees to focus on other high-priority tasks resulting in an increase of productivity.

Factory workers will be seeing more of robotic technology as according to the International Federation of Robotics, by 2020 more than 1.7 million new industrial robots will be installed in factories worldwide. The automotive industry is still the major customer of industrial robots with a share of 35% of the total supply in 2016.

The other industries that we believe will see an increase in use of robotic hands are manufacturing (within the assembly line for quality control purposes or dealing with goods that may be unsafe for human handling), healthcare (service robots that could assist people, such as the elderly or those suffering from dementia, with tasks such as preparing meals and dressing or within biomedical and pharma, preparing vaccines and dispensing medicines) and nuclear decommissioning (using robots to clean up nuclear sites where hazardous materials make it risky for humans to go in).

In mass production factories robotic arms are not a new thing, rather a necessity, because it’s a very good tool for repeating the same action, but what about workshops where every product is unique?

We’ve identified that many robotic grippers and arms on the market don’t have the technology needed to grasp unknown objects successfully or alternate between grasping different product parts, so while it may perform a repeated action very well, it doesn’t fulfill the need when another object gets thrown into the mix. That’s one of the fundamental reasons as to why we’re developing our Smart Grasping System which uses our Modular Grasper as a key component. It’s an intelligent robotic grasper for manufacturing and industrial purposes that can be utilized in companies and workshops that deal with different types of products.

Using our expert skills in grasping and manipulation, we’ve created a next-generation tool which has a library of grasps allowing it to grasp many different types of objects using just one hand. It has torque sensing and a vision system for an accurate and reliable grasp reducing high grasp-failure rates. It saves time and money on buying different types of robotic technology to suit different products and because it’s robust and reliable, it also saves on maintenance and repair costs. I guess you can say the future of the robotic arm in performing at its best with a range of objects is next and upcoming.

It was deeper than a love for technology, we wanted to excel at grasping and manipulation for robotics and by evolving our skills and knowledge in that area, we became a team of experts.

In the 90’s you with the group of other hobbyists decided that building a robotic hand would be a great starting point for a company and as it now is obvious – it turned out to be a success. Let’s imagine that same situation but in 2018. What new engineering innovation you’d believe to be a good starting point for a new business today?
I believe the good starting point for any new business or enthusiast is to create something that the market lacks or hasn’t advanced in yet. Back in the 90’s, we knew we wanted to set ourselves apart and offer something of maximum value, so we challenged ourselves to build one of the most complex and essential natural engineering parts of the human anatomy: the hand. It didn’t happen overnight and wouldn’t have happened at all had we not been passionate about what we were doing. It was deeper than a love for technology, we wanted to excel at grasping and manipulation for robotics and by evolving our skills and knowledge in that area, we became a team of experts. Now that robotics has got much more recognition today, there’s a keen interest in using it to help with real-world problems such as the ageing population crisis or nuclear decommissioning. So, any new engineering innovations geared to solving pressing issues within society is a great start.

---

About the robotic hand
The Shadow Robot Company developed a humanoid robot hand system called “The Shadow Dexterous Hand”. The hand is comparable to a human hand in size and shape and reproduces all its degrees of freedom. The system is commercially available in pneumatic- and electric-actuated models and currently used in a wide range of institutions including NASA, Universities of England, EU research projects such as HANDLE.