KEY FACTOR MOVEMENT PRIMITIVES: HOW CAN WE MAKE A ROBOT LEARN FROM HUMANS?
HOW DOES A ROBOT LEARN HOW TO PERFORM A TASK NOWADAYS?
Since Henry Ford’s assembly lines, automation systems have been developed and integrated in the industry all over the world. Clear examples of this are Amazon’s automated warehouses or Coca-Cola’s dark sheds.
While these examples are stunning, it is also true that nowadays, these robots need their own warehouses where no interaction with humans happens. Also, high qualified robotics engineers have to program them to perform the required repetitive tasks.
In the future, we imagine a world where anybody, without any prior knowledge in robotics, could buy a robot, set it in their workshop and start teaching the robot to perform some task interacting with the human workers. But, what should the robot learn? Welcome to the Movement Primitives.
WHAT ARE THE MOVEMENT PRIMITIVES?
Movement Primitives are the set models that represent the different type of actions the robot could do.These models will generate trajectories the robot should follow in order to perform a certain task.
Let’s assume we want a robot to learn to pick a certain object. Instead of programming the task, we can make the robot learn from human demonstrations. A human could move manually the robot arm to the object we want to pick several times while changing the object position and the robot will learn a model that represents that action. Afterwards, if the object is set in a completely new position, the robot will be able to generate a trajectory to pick this object.
MOVEMENT PRIMITIVES IN SHAREWORK
SHAREWORK project benefits from the use of Movement Primitives for task recognition and also prediction of human movement. Movement Primitives can represent the different type of motions the human does and so, they could be integrated as an additional feature for a task classification module. On the other side, Movement Primitives can provide a probabilistic prediction model about where the human is going to be, in order to improve the robot motion planners.
During the SHAREWORK project, Context-Aware Movement Primitive Segmentation algorithms are going to be applied. The algorithm will be able to segment the recorded human motion online and learn the simplest blocks of which the human motion is built in order to solve a complex task.
SHAREWORK project will use Movement Primitives for learning from human actions by understanding the worker’s behaviour and anticipating tasks
These learned motion primitives are contextualised with the objects in the environment in order to adapt to the different positions the same object can be placed. Afterwards, these Movement Primitives will be used as the building blocks of the task classification algorithms.
These same Context-Aware Movement Primitives will provide a probabilistic prediction of the position of the human in the future. Mixed with robot motion planner, the robot will plan safer trajectories, avoiding the collision with the humans.
ROBOTICS INTO HANDWORK MANUFACTURING: THE FIGURE OF THE TECHNOLOGY COACH FOR LEADING AN EFFECTIVE AND SMOOTH IMPLEMENTATION
THE NEW CHALLENGE OF ROBOTICS: TAKE OVER THE HANDWORK MANUFACTURING
The Robotic Industry is growing bigger day by day, launching on the market several advanced solutions which promise to revolutionise the way factories produce goods and assembly parts in a near future. As a result, also the expectations of Industry are constantly increasing and they look at robots as the definitive solution to save time (and, of course, money) and boost profits of their manufacturing companies.
The handwork manufacture is the new battlefield that robotics has begun to conquest in recent years, but despite this there is still a lot of room for improvement and expansion. In fact, while nowadays almost all the operators performing repetitive and simple tasks have been replaced with highly automated assembly lines, the same has not yet happened for handcraft. This is mainly due by the fact that manual labour requires such high skills and a human sensitivity that, at the moment, a robot cannot replace the operator in almost all the cases.
The automation of assembly lines based mainly on handwork are quite a challenge for the robotic industry due to the need of high skills and sensitivity
The solution that robotic producers have found to overcome this limit is the launch on the market of collaborative robots (also called cobots). Cobots aim to support handcrafts in the most repetitive, heaviest and uncomfortable tasks rather than completely replace them, allowing to preserve the high quality generated by manual labour and speeding up the productivity at the same time.
Unfortunately, the implementation of a robotic system into the production process is everything except easy and instantaneous and the benefits the robot could bring are not so obvious. In fact, the selection of proper components and a smooth integration are crucial to generate the positive impact on productivity, quality and operators’ working conditions which factory owners wish. And this is even more true for the implementation of collaborative robots into handwork manufacturing, where the system should reach high performance without neglecting the operator safety and the human acceptance.
Collaborative robots offer handcraft industries a support in repetitive and uncomfortable tasks for operators
Indeed, there is the need of an intermediary figure which could drive the end-user through the jungle of robotic products and serve as an interface between the salesman approach of suppliers and the concreteness of craftsmen undertaking in the workshop. This figure within Sharework project is commonly named as “Technology Coach”, and STAM is in charge of filling that role for one of the four use-cases, namely the assembly of rotary table in Goizper factory.
THE TECHNOLOGY COACH: NOT ONLY A MATTER OF DESIGN
If the role of the Technology Coach is essential in the implementation of a robotic system into production lines, it could be even crucial when we talk about the introduction of cobots supporting handcrafts because of the aforementioned reasons. But exactly, who the Technology Coach is? And what it is supposed to do?
Generally speaking, it is a work team with well-mixed competences which belongs to a high-tech company or a R&D center. Unlike suppliers or system integrators, the Technology Coach not only proposes the design of a tailored robotic cell, but they follow a holistic approach which involves the end-user to reinvent and redesign the workflow and the workshop layout of tomorrow in a better way.
The Technology Coach should listen and understand the needs and the objectives of the end-user. Satisfying these wishes and achieving these goals will be its main mission. Once an agreement is found, Technology Coach should select suitable commercial solution, get in touch with suppliers, design the robotic system, reinvent the workflow and reshape the layout if needed, install and test the solution.They should take care of the whole process of implementation of the robot into the shop floor, from the definition of requirements to the commissioning and final test. This is an iterative and continuous process, some steps should be done several times until finding an equilibrium among the end-user expectations, the current robotic supply, the budget, the operator acceptance and well-being, the timing and so on.
Technology Coaches help industries in the implementation of new robotic systems, leading the whole innovation process matching he end-user’s demand in the most effective way
Most of the work of a Technology Coach can be done in its headquarters, using modern communication network to periodically update the end-user about the work progress. However, it is highly recommended to have a face-to-face with the end-user at the begin of the project, as well as to visit the shop floor and talk with operators working there to have a clear picture of the process.
The value added to the figure of the Technology Coach is to put at the stake all its set of cross-cutting competences and resources to satisfy the wishes of the end-user. Technical Coaching does not simply mean designing a robotic cell, but leading the whole innovation process of the manufacture, matching the robotic supply to the end-user’s demand in the most effective and smooth way.
THE SHAREWORK EXAMPLE: A COBOT FOR SUPPORTING THE MANUAL ASSEMBLY OF ROTARY TABLE
As mentioned before, STAM embodies the role of Technology Coach in Sharework project to support Goizper and coordinate partners contributions in the implementation of a collaborative robot system in their servo rotary table assembly area. As the assembly of these components requires a remarkable human sensitivity and manual skill, the main aim of Goizper is to delegate to the Sharework system the most heavy and repetitive tasks of the assembly, while preserving worker focus on the most skilled operations.
As first step, we worked together to select the most suitable tasks to be delegated to the cobot and to define in details each scenario. In order to have a complete and clear picture of the workshop and the process, we carried out a technical visit to the factory together with the consortium. We had the chance to directly experience the working conditions (e.g. spaces, loads, noise), as well as to talk in first person with the workers to collect their feedback.
Once the use-case was pretty defined, STAM began to serve as an interface among the end-user and the external world. Our role was double: on one hand we translated the requests of Goizper into requirements and clear specification to effectively communicate with commercial suppliers and with the multiple software developers involved in Sharework, each one focused on different specific tasks. And, vice versa, we collected feedbacks, requests and suggestions from all the different entities involved in the Sharework project to provide a unified vision to Goizper and to propose them a feasible human-robot collaborative solution that could meet their expectations.
For these purposes, 3D models and views (see the following pictures), as well as all the other visual medias, are very useful to finally put on paper the developed concept and allow other partners to understand it in an easy and clear manner.
It has been a long and iterative decisional process, but, in the end, we did it! We reached a final solution reinventing the rotary table assembly workflow to enhance the productivity, exalt the qualities of workers and safeguard their well-being.
However, our job is just starting out: we have to work hard in the next months to make this design on paper a concrete reality. Stay tuned!
TRAVELLING TO GOIZPER 4.0 WITH COLLABORATIVE ROBOTICS ON BOARD
During the last years, Goizper’s mind has changed little by little, complementing new technology developments with the power transmission components portfolio. The obtained good results have certified that new technologies integration is the correct way to follow.
A STEP FORWARD BECOMING GOIZPER 4.0
Goizper S. Coop. has already integrated several Industry 4.0 solutions related to its market products, such as smart-G, an IoT monitoring system integrated on mechanical components in order to predict failures. However, now Goizper has started a new era introducing Industry 4.0 solutions on its own shop floor in Antzuola, Basque Country (Spain).
On the one side, new machinery, such as milling and turning machines, are running in Goizper’s shop floor, with integrated IoT solutions and direct connection with the supplier (machine experts) through internet. On the other side, Goizper has decided to invest in new technologies on the manual assembly area, where operators’ ergonomic issues concern the company.
Conventional robots have never fit in the assembly area due to the flexibility degree that Goizper’s products assembly require and the high variety and low quantities of each reference. However, Goizper decided to go further and participate in Sharework project as we believe that collaborative robots will be able to provide the flexibility that Goizper’s assembly area needs. Within the project, Goizper Industrial will validate a new design of their workplace to introduce collaborative robots in their servo rotary table assembly area.
Several benefits are expected from Sharework project, including an improvement of ergonomics in the workplace, avoid work-related injuries, improve the final quality of the product and a significant reduction of the assembly process’ cycle time.
OPERATORS’ ERGONOMICS CONDITIONS TO IMPROVE
A new way of introducing robotics into the workplace has risen with the development of collaborative robotics (Cobots), which allow the operator to share the assembly area with robots without the need to be separated by fences. Managers in Goizper think that Cobots introduction in the assembly line by means of Sharework project, will improve operators´ ergonomics conditions.
Power transmission components’ assembly, such as turning tables, request many manual adjustments applied by human force. Hence, some tasks could lead to back injuries or muscular pain in operators.
OPERATORS AND COBOTS, A SUCCESSFUL TEAM TO RISE PRODUCT QUALITY
For sure, with the introduction of cobots for operators’ assistance, the quality of the assembled product will be higher. Obviously, automated processes provide a higher percentage of good parts compared to manual processes. In Sharework industrial scenario, bolts will be locked with the same torque in a repetitive and more precise tool than nowadays.
Goizper is eager to see how much the quality of their products is improved after cobots incorporation, during Sharework project.
SPEEDING UP THE ASSEMBLY PROCESS
Productivity is not the main goal of this technology integration; nevertheless, assembly cycle time is expected to be reduced due to the fast execution of UR10 Cobots. Within Sharework scenario, which will be applied to our servo rotary table assembly area, time reduction must provide a better lead time of the product and economic benefits.
THE SAFETY OF OUR EMPLOYEES AT THE CENTER OF ALL THE OPERATION
As none of these new technology’s integration could run before ensuring human safety and make sure that all safety requirements in the shop floor are met, Goizper’s Human Resources Department – the bridge between the new technology and the human – is also closely participating in Sharework project.
Goizper, as one of Sharework’s project industrial scenarios, aims to use the collaboration robotics technology for assisting and helping operators in their daily tasks, such as handling heavy parts to the operator and passing them the correct part to be assembled or the tool to be used at the correct time. Under no circumstances the operator will be replaced by the collaborative robot but just another tool available to help for the operator, in addition to other technologies or devices already deployed in the assembly workplace.
CEMBRE AND THE SHAREWORK PROJECT: A STORY ON COLLABORATIVE ROBOTS’ DEPLOYMENT IN FLEXIBLE MANUFACTURING ASSEMBLY AREAS
The manufacturing industry is observing great changes in these years. Its radical evolution is due to the growing demand of the market, accompanied with an increase of variety. Both these characteristics force us to gain more flexibility, accuracy and efficiency in our production departments, together with a continuous cost reduction.
The industrial manufacturing applications have seen the introduction of a new category of robots – collaborative robots (or “cobots”) – designed to physically interact with humans in a shared environment, without the typical barriers or protective cages used in traditional robotic systems.
Our vision in Cembre S.p.A. is to take advantage from cobots and from the smart manufacturing as a whole in order to increase the service rate and improve the worker’s ergonomics. We think that human-robot collaboration (HRC) is prime for increasing both these factors and that Sharework EU project is fundamental for developing more knowledge about HRC, helping filling the gap between the robotic and the human worlds.
Cembre could have gone to full shop floor automation but we would have lost human flexibility. Collaborative robots provide a half way solution between the human and robotics world
THE INTRODUCTION OF COLLABORATIVE ROBOTICS IN ADVANCED MACHINE TENDING
Cembre S.p.A., based in Brescia (Italy), is always eager to employ new methods or technologies that can support the development and the improvement of its entire production system. For this reason, the company constantly collaborates with universities, start-ups, research centres and others associations.
The Sharework project perfectly fits in this approach, since it will be pioneering in the employment of HRC using collaborative robots within manufacturing systems and, in general, in industrial operations.
The objectives we want to achieve are many.
- Gain flexibility in managing the machining of a large variety of parts without the need to develop expensive equipment and plan a long training for every new product that enters the shop floor.
- Employ our operators in more value-added tasks, such as inspections and process optimisation, and move non-value-added activities and repetitive task, such as motion or transportation, to the cobots.
- Increase operators’ ergonomics, which itself is one of our primary targets together with operator’s welfare and satisfaction. These latter should increase because the skills required to an operator will be brought to a higher level thus improving the interest for the job inside machining shop floor.
- In the long term, once the complete system will be deployed on the whole shop floor, we also expect a decrease in lead-time and quality losses.
SHAREWORK PROJECT DEPLOYMENT AT CEMBRE’S PLANT
We are currently implementing the Sharework project in the machining shop floor located in Brescia’s plant: production engineers, human resources and operators are all engaged. Most of them are new to this kind of projects and these technologies, aspects that makes Sharework especially challenging for the company. Regarding this, the contributions of the CNR, the Italian research centre, and of MCM, our technologic partner, are very important to us.
The role of people is fundamental to create deep connections between working conditions and technologies. In fact, one of the main challenges for the future is to understand how this new paradigm for the manufacturing systems could generate benefits to the entire production chain. Who if not humans could imagine this?