Final Report Summary - HARCO (Hierarchical and Adaptive smaRt COmponents for precision production systems application)
Future machine manufacturing will be based on adaptive structures — systems of machines and robotics with high degrees of autonomy and intelligence. However, further developments are needed to create affordable components that are extremely stiff, light and dampened, and controlling systems based on measuring systems and sensors.
An EU-funded research project, 'Hierarchical and adaptive smart components for precision production systems application' (HARCO), aimed to design a new class of smart components suitable for a wide array of functions for machine tool applications.
Involving project partners from research institutes and industry from six countries, the HARCO team created a hierarchical structure of units with a highly integrated control logic. This can be adapted to different operating environments and tasks.
Work initially focused on testing sensors, actuators and smart materials such as magneto-rheological fluids and magnetic shape–memory materials. The project then created several working demonstrators, including a modular and scaleable milling machine, parallel kinematics machine (PKM) robots, and a serial robot with an active wrist that increases the robot's stiffness.
The team continued to test their prototypes until the end of the project in June 2013. The project is laying the groundwork for a revolution in manufacturing machinery, helping to improve the efficiency and competitiveness of European industry in the 21st century.
Project Context and Objectives:
The HARCO project was given the go-ahead by the European Commission in July 2010.
Adaptive structures will be at the frontier of knowledge and will revolutionize machine tool and manufacturing machinery design and construction in the 21st century.
The challenge in this area is to realise availability of “extremely” stiff, light and well damped structures with fully and deeply integrated new adaptronic devices based on electromechanical and electronic devices, measuring systems, sensors and actuators. Getting more intelligent and integrated structural solutions in a cost-effective way is essential to meet performance targets in commercially viable machines and really introduce enormous benefits in machine tool design and development.
Therefore the primary goal of HARCO is to achieve cost-effective structural solutions consisting of a new class of Smart Components (belonging to machine tools applications) based on plug-and-produce “Modular Adaptronic Devices” which integrate smart and multifunctional actuators/sensors capable of performing a wide array of multiple functions, ranging from high and adaptable damping and stiffness characteristics to more demanding new requirements, such as active structural measurement and control function to achieve extremely high dynamic/thermal stability required in fast and precision machining.
The approach followed by HARCO is the hierarchical combination of lower level units named “Functional Bricks” to generate higher level modules called “Adaptronic Modules” which in turn are used and integrated into machine parts to generate the master “Adaptive Smart Components” or ASC.
Then the basic idea is to design and develop a sort of “fractal” and “hierarchical” elements (not only mechanical hardware but also controllers and software) that can be easily put together (plugged-in) to form/produce higher level modules/components (modules that build modules!) for active vibration control, thermal compensation and adaptive fixturing in precision machine tools applications.
The ASC is an “intelligent” structure which contains highly integrated control logic and electronics that provide the cognitive element of a distributed or hierarchic control architecture (high level control link to machine CNC), which enables the changing of structural properties and/or characteristics to properly adapt the structure behaviour itself to a specific operative/environmental condition.
In the first half of the project, great effort has been devoted to investigate and characterize the candidate sensors, actuators and materials for Functional Bricks and Adaptronic modules.
In particular specific test benches have been designed and set up in order to verify the robustness and the duration of mechatronics components.
For instance, highly relevant work has been performed in order to better understand the potential of piezo stack actuators in terms of repeated cycle loads and integrate the limited data available from the manufacturers and from scientific literature.
This activities increased the confidence on piezo stack actuators applicability in industrial environment and justified their adoption in adaptronic modules.
Similarly, research on smart materials, such as magneto reological fluids or magnetic shape memory materials has been carried on.
In parallel, the target performances and the concept design of functional bricks and adaptronic modules took place, such as the virtual validation.
In the second half of the project, after a full performance characterization, the Adaptronic Modules found their final collocation in the five HARCO demonstrators, accordingly to various manufacturing scenarios.
The selected modules have been interfaced with existing industrial equipments such as milling machine tools, parallel kinematic robots and serial robots.
Demonstrator #1 – High speed milling machine
The main requirement of the Demonstrator #1 is to demonstrate that a conventional High Speed Milling Machine (HSM), designed for performing standard 3-axis machining (mid size mould and dies manufacturing, aerospace parts…) can be adapted and improved, equipping it with Plug&Produce (P&P) components developed according the HARCO approach. Specifically, a DL155 FIDIA HSM , has been equipped with the Plug and Produce two-axes micropositioning table designed by CeSI and MACH4Lab and manufactured within the HARCO project, with the aim of completing the machine with two ultra-precise additive axes (namely U axis and V axis, which are parallel to X and Y machine axes). The UV micropositioning table turns the DL155 in a HSM capable of performing close-to-micromachining operations, i.e. machining of meso-to-micro details (< 10 mm, > 0,001 mm). Adding close-to-micromachining capabilities to a conventional HSM is desirable in order to fulfill the gap between micro-machine tools (unable to handle workpieces larger than 800x600x600 mm and unfit for performing roughing and semi-finishing operations) and conventional HSM (unable to achieve roughness and edge angles typical of micro-machining). A typical field of application is represented by the automotive lighting sector, where designers and lighting technologies pushes mould manufacturers to guarantee micro-machining precision on large parts.
Demonstrator #2 – Smart system implemented in a high speed PKM robot for pick&place
Conceived for high-speed and high-accuracy pick-and-place tasks, the parallel robot Par2 aims at reducing industrial operation cycle times. As a consequence of its high acceleration levels, the end-effector precision at the stop positions is subjected to undesirable vibrations of the manipulator’s arms, leading consequently to an increase in the cycle time.
In order to increase the accuracy and the productivity of the parallel kinematic robots, TECNALIA and the other partners have followed two strategies. On one hand, we have developed and implemented an adaptive controller, such as, computed torque control. On the other hand, we have developed and implemented a smart device to damp the vibrations of the mobile platform of the robot in the placing position. In order to damp vibrations originated by the frame of the robot and by robot’s arms, HARCO approach has been to implement an active device to damp the vibrations of the robot frame and a smart device to damp the vibrations of the robot itself (active and passive solutions could be included if needed).
At the beginning of HARCO project was studied the influence of the flexibility and inertia of the PAR2 structure in the robot positioning. In order to compensate the robot inertia, it was needed to develop adaptive control algorithms to reduce the position-tracking error measured by the robot systems. Three approaches have been proposed and tested: Cartesian control, Cartesian control with FFW compensation and Computed Torque controller, finally achieving a relevant reduction of tracking error.
Demonstrator #3 – Smart system implemented in a serial robot
The main disadvantages of serial robots are their low stiffness and their low accuracy, disadvantages that make difficult their use for machining tougher materials or medium-soft materials with high accuracy.
Therefore, the objective is to reduce the vibrations in the robot’s wrist to increase its dynamic stiffness and thus, enable them to machine tougher materials, to machine medium-soft materials with high accuracy and/or to machine soft materials with higher productivity. In particular, this activity aims at increasing the dynamic stiffness of a serial robot to enable it to perform most of the operations required in the aeronautical sector: drilling, riveting, milling of boxes and contouring.
In order to fulfill this objective an active damping device has been developed and validated.
Demonstrator #4 – Planar PKM
The planar PKM demonstrator was intended as a proof of principle for the effectiveness of add-on adaptronic module (the Smart Joint) in achieving active vibration suppression on a state-of-the art industrial motion system.
The results achieved confirm the validity of the design procedures, leading to an original, low-cost realization of a high-performance planar PKM, and the modularity and effectiveness of the Smart Joint in achieving active vibration suppression.
Demonstrator #5 – MT with advanced Thermal deformation control
Thermal errors of machine tools, caused by internal and external heat sources, are one of the main factors affecting CNC machine tool accuracy. Internal heat sources include the five-axis drive motors, bearings, friction in the ballscrew nut and support bearings, electrical components, and other similar effects. Among these sources, the power loss of the motors is one of the most important internal heat sources. External heat sources are attributed to the environment in which the machine is located, such factoring heating or air handling systems, neighbouring machines, opening/closing of machine shop doors, variation of the environmental temperature during the day and night cycle and differing behaviour between seasons. The complex thermal behaviour of a machine is created by interaction between these different heat sources. The error compensation system is often considered to be a more economical method of decreasing these thermal errors.
The Structural Monitoring Module developed by University of Huddersfield with the consortium support has been applied as a whole to two elements on a single machine tool. Further validation is required over a longer period and also on a different machine type. Various Artificial Neural Networks models have been trialled during this project, with promising results. Further development by increasing the number of sensor inputs to determine any benefit is an interesting avenue for research.
Commercialisation of the system would require industrial hardening of some of the technology. That said, the system could be trialled in a manufacturing environment as it stands, since the robustness has been proved during the project.
Future relationship and collaboration among partners
Many of the results will need further development, certification/standardisation costs or marketing related costs. The amount of resources will vary depending on the result, and tends to be proportional to the level of aggregation in the modularity.
In some of the cases, the cost/resources required could be in the order of magnitude of several hundred thousand €. Such an investment requires senior management approval, and is not a straightforward proposition. This means that the inputs for the decision making will have to be available, beyond the technical level, very likely in business terms. Therefore, aspects such as the cost/pricing elements, and market related inputs will need to be covered to maximize the probability of getting approval for further investments.
The fact that there will be further resources required for the future implies that efforts will be required not only individually, but at collaborative level. In the case of future collaboration, the contribution from several partners would be necessary. When needed, partners have defined collaboration agreements to go on working together once the HARCO project ends with the aim of guaranteing
that the demonstrators can be developed in full.
Development of a supply chain of industrial grade:
The implementation of the different modular results into the modules or machines will require defining a supply chain. This raises several questions such as how each partner can become part of the supply chain.
This question is particularly important for the research/university partners.
They are actually interested of becoming commercial partners, not only in terms of IPR transfer, but in terms of the actual component production.
Individual or joint exploitation of results
Given the modular approach of the project, in which results are developed at different levels, in a modular fashion, exploitation can also be done modularly, at the Functional Brick level or even at a lower level (e.g. sensors and actuators).
In the case of some of the Adaptronic Modules, individual exploitation can take place at the level of design or retrofitting.
List of Websites: