One of the key issues in robotics is co-ordinating data so that control functions work effectively. At HBK, our expertise lies in sensors that measure torque, force, mass, pressure. But we also know that other data dimensions are involved: acceleration, contact, distance, gyroscope, humidity, inertia, light, navigation, position, pressure, proximity, sound, temperature, tilt, voltage – and more.
Custom Sensor Assemblies for Robotic and Cobotic OEMs
Why do we have robots? A lot of good reasons spring to mind. They can be grouped around activities, environments, productivity, and economics.
Some activities are repetitive and boring. Tightening the wheel nuts may well be an essential step on a car production line, but its not an activity that can be classified as a “fulfilling career option”. At the other end of the scale, there are activities that are complex and difficult. But add the element of repetition back in, and the frustration returns. How much better to delegate repetitive tasks such as machining and milling to CNC machines and set the human imagination free to experiment with new ideas.
There are many work environments that people find unpleasant. Places where there is too much heat or cold, humidity or aridity, noise or vibration, for example. Some environments pose an inherent health risk: paint booths, nuclear reactors, volcanoes, polar regions.
And then there are work environments that are downright dangerous: mines, deep sea operations, higher altitudes, outer space. Instead of risking life and limb, we can build robots to handle such environments.
Productivity is a great reason for having robots. Machines can perform work faster, more accurately, with greater consistency than humans. And they require less downtime. Taken collectively, these ideas add up to a strong economic argument for robots. Robots that do work that people don’t want to do, and produce greater value than the costs of ownership and operation, are a benefit to society.
There is one other reason why we have robots: human curiosity. The simple fact is, some people get an enormous kick out of designing elegant solutions to sophisticated problems.
Robots as we know them today are a combination of computer science and engineering. Sensors gather data about the environment; a controlling program determines how and when the robot will act; actuators implement the action; sensors gather data about the interaction and provide feedback to the controlling programme, Repeat infinitely, in real time.
That simple phrase – ‘sensors gather data’ - hides enormous complexity. Each sensor must observe the signals it was designed for, reliably and accurately. But control algorithms are seldom about just one piece of data - multiple signals must be transformed into actionable data and forwarded to the controller. At HBK, our expertise lies in sensors that measure torque, force, mass, pressure. But we also know that other data dimensions are involved: acceleration, contact, distance, gyroscope, humidity, inertia, light, navigation, position, pressure, proximity, sound, temperature, tilt, voltage – and more. One of the key issues in robotics is co-ordinating data so that control functions work effectively.
The issue of sensing and programming becomes more complex the moment humans come near. The idea of a robot as a helper – a Co-Bot (collaborative robot) - introduces additional safety parameters. Robots must not move with a torque that would injure a person, so sensors must react faster, actuators must decelerate more quickly. These aspects, too, have been discussed and agreed internationally: IEEE as well as ISO have issued standards.
The issues become even more complex when the autonomous robot – controlled by an Artifial Intelligence (AI) program - ventures into public life. For although it must act independently of its programmers, the robot may in practice inherit assumptions and biases that reflect the programming teams’ definition of ‘normal’, ‘acceptable’ or ‘desirable’ decisions. Leading institutions such as the IEEE, as well as universities like Stanford and MIT, now recognise ‘Robot Ethics’ as an important field.
For all their sophistication, static production line robots are conceptually ‘simple’. They typically perform one specific function: for example: cut, press, weld, or paint. The individual robot does not know the function it performs; nor the process that precedes or follows it; much less the concept of teamwork.
The real intelligence lies in defining the sequence of robotic actions, which is provided by human experts. And ‘finding a better solution’ is an innately human activity. The search for a more effective conceptual model for a robot is a fundamental issue. It impacts not just robot design but also practical operation. Interestingly, many models come from nature.
Behaviours observed among insects, for example, contribute to the controls for fetch-and-carry warehouse robots. Each individual robot follows simple rules to ensure it achieves its mission without getting in anothers’ way. The more sophisticated conceptual models use insights from observing swarms in nature – of insects, birds or fish – to enable individual robots to share information and co-ordinate their actions, for increased effectiveness.
The market potential for production line robots is constantly expanding. Robots that move stock around the factory floor are mainstream. In the search for increased efficiency, some vendors provide the ability to climb the shelves when picking stock; other have re-designed both the storage system and the robot that accesses it, to optimise use of warehouse space.
Follower robots can carry heavy loads (like burro for farmers, or gita for city dwellers). Autonomous robots are gaining traction – both on private industrial premises (like factories or mines or for cleaning warehouses) as well as public spaces, for activities such as delivery services. Meantime, other concepts have appeared and created additional opportunities such as mobile robots, aerial robots and soft robots.
Will robots transform the world of work? They already have. Many activities that were repetitive, boring, uncomfortable or downright dangerous are already handled by robots. Meantime, human resources have been freed up, enabling people to do what they do best: create sophisticated and innovative solutions to challenging problems. It looks like that trend is set to continue. If we can increase productivity and bring about a more even distribution of wealth both within and between the different societies across the globe, a world with robots can definitely be a better place.