The Sweeper consortium was invited to hold the first live demonstration of its new sweet pepper harvesting robot at the De Tuindershoek greenhouse horticulture firm in IJsselmuiden. The so-called ‘Sweeper robot’ is the world’s first harvesting robot for sweet peppers to be demonstrated in a commercial greenhouse. An audience of over 40 interested parties watched the harvesting robot pick its first commercially-grown sweet peppers.
The Sweeper robot was designed to harvest sweet peppers in a cultivation system based on single plant stalks in a row, a crop without clusters and in little foliage near the fruits.
In earlier test set-ups in a commercial greenhouse with a V-type double-row cultivation system the harvesting robot achieved a harvesting percentage of 62%. Based on these test results, the Sweeper consortium expects to be able to bring the commercial sweet pepper harvesting robot to the market in about four or five years.
Further research required
Until then, further research will be needed to enable the robots to work faster and achieve a higher success percentage. Additionally, commercially viable cultivation systems must be developed that are more suitable to the robotic harvesting of crops. The test and research results are not only suitable for the automatic harvesting of sweet peppers; the data can also be used to robotise the harvesting of other crops.
International research partnership
Sweeper is a partnership between Wageningen University & Research (WUR), sweet pepper farm De Tuindershoek BV, the Umea University in Sweden, the Ben-Gurion University in Israel, the Research Station for Vegetable Cultivation and Bogaerts Greenhouse Logistics in Belgium. The study receives financial support from the EU’s Horizon 2020 programme and is also funded by the Dutch Horticulture and Propagation Materials Top Sector.
Successor of CROPS
The Sweeper robot is the successor of CROPS (Clever Robots for Crops), an EU project launched by WUR, in which WUR and the other participants developed a robot that can make a distinction between a sweet pepper plant’s fruit, leaves, stalks and main stems. As a result, the robot can harvest sweet peppers without damaging the fruit, leaves, stalks or stems.
Source and photo: www.sweeper-robot.eu. Video: Wageningen UR greenhouse horticulture.
A consortium of three companies, Ecoation, Metazet FormFlex and Micothon, has launched a new robot for scouting plants in greenhouses.
Revolutionary patented sensors register plant diseases and plant production factors before human eyes can discover them. The data generated is automatically displayed on a greenhouse map. This finally makes it possible to steer and protect the plants on an almost individual level, leading to minimal need for crop protection combined with maximum production of “always healthy plants”.
The robot registers the diseases and development of the plants in greenhouses with technology based on the patented Saber sensor, which detects pests, diseases and deficiencies at an early stage. It can also be equipped with cameras, a gyro sensor and sensors for RH%, temperature, CO2, crop top temperature and a PAR sensor, resulting in almost total plant control.
Stand number: 12.107
Thanks to the unique pull wire installation robot, working at great heights and installation faults are no longer an issue.
Nowadays no greenhouse is built without a screen cloth. More and more greenhouses have double or triple screens for energy saving, sun shading, blacking out or reducing light emissions. Thanks to the innovative slip-in systems, installing screens is already much safer and easier than it was ten years ago.
Problem solving robot
On the other hand, a lot of dangerous work is still being done at heights of 6 or 7 metres. For every hectare of greenhouse, 2,000 metres of pull wire have to be installed. Installing pull wires is normally done manually, and as the wires have to be installed between thousands of upper and lower polyester wires, installation faults are easily made. With the innovative robot from Van der Valk Horti Systems, these problems are solved.
Stand number: 12.223
The ISO Robot Plug Planting Machine is a machine with robot arms that can plant small plants in the ground from a plug tray. This method of planting plugs in the ground with robot arms is completely unique anywhere in the world.
The current machine, which is specially designed for Lisianthus plants, can handle 6,000 plants per robot arm, and the maximum capacity can be increased to 18,000 plants per hour with a third arm. It can all be operated by one person. Up to now, the plants have been planted out by hand. With the humidity and warmth in the greenhouse needed to grow Lisianthus, this is a very tough, labour-intensive job. The machine is designed to enable the operator to control the plants as they are being planted.
Stand number: 11.307
The new smart packing line from WPS consists of five individual modules, all members of the SmartStaff robot family. Modules can be combined based on the grower’s specific needs.
Smart Picking is designed especially for companies with a capacity of more than 2,500 plants per hour. It picks up plants from containers and places them in plant carriers on conveyor belts. The labelling module has a capacity of more than 2,000 plants per hour. Special feature: pots are labelled with in-line printing directly from the OCS software. The third module is for destacking trays and works with a wide range of trays which can all be learned by the OCS software.
Placing plants without damaging them is the speciality of Smart Placing. It has a capacity of more than 1,500 plants per hour. Finally, Smart Wrapping (> 200 trays per hour) wraps trays filled with plants. This module reduces the amount of plastic used in the plant packing process.
Stand number: 11.201
At the coming World Trade Fair for Horticulture at IPM in Essen, scheduled for 23 up to and including 26 January 2018, the ISO Group will unveil a new machine. This machine is a combination of the already well-known ISO Cutting Planter 2500 (automatic cutting planter machine) and the ISO Vision Planter (transplanter).
The Cutting Planter 2500 is now a well-known name in worldwide horticulture. Globally, more than 80 of these ISO Cutting Planter 2500 machines are in the field. Considering the fact that many nurseries have a limited season for sticking cuttings, ISO Group has developed a machine that, in addition to the cutting planting, also has the ability to transplant rooted plants.
Through the easily interchangeable head it is possible to use the machine as a transplanting machine after the cutting planting season. Because the gripper head is interchangeable, it is also possible to process different sized plants and different tray and pot sizes. For instance, in consultation with the ISO Group, it is possible to design one or more heads with 1, 2 or multiple grippers, which, depending on plant and pot size, can ensure the right plant handling and speed.
In this way, the machine can be used for a longer period of time, in multiple seasons and for multiple plant varieties. Besides the ISO Cutting and TransPlanter, the ISO Cutting Planter 2500 will also be at the show at IPM in Essen. The ISO Cutting Planter 2500 makes it possible to carry out cutting planting automatically. The ISO Cutting Planter 2500 is operational at multiple reputable grower locations worldwide and is capable of planting many types of cuttings automatically.
Photo: ISO Group.
Orchid nursery Hazeu Orchids in Delfgauw in the Netherlands has robotised part of its internal transport system. The robot sets down pots with young orchid plants on mobile benches, where they are grown on until they are ready for sale. The art is to get as many orchids as possible on each bench so as to optimise productivity at the nursery. But to avoid damaging the plants, it is important not to put the pots too close together.
Javo, a Dutch systems engineering firm from Noordwijkerhout, has built Hazeu Orchids a robot on the Sysmac platform with electronic components from Hoofddorp-based Omron. This robot sets down around 2,000 to 3,000 plants per hour on the mobile benches exactly the right distance apart.
The family business in Delfgauw grows various Phalaenopsis varieties on seven hectares. Most of its 4.5 million plants go directly to large retail chains – supermarkets, hardware stores and garden centres – without going to auction first. According to co-owner Chiel Hazeu, who is responsible for cultivation and technology, growing Phalaenopsis is a delicate process. “The orchid is a fragile plant. But retailers want beautiful plants of consistent, uniform quality and, of course, attractively priced. An orchid won’t sell if the quality isn’t up to scratch. This not only places tough demands on handling in the nursery, during transportation, at the logistics centres and in the stores but also on the packaging and presentation.”
Plants per square metre
Besides temperature, atmospheric humidity and fertilisation, one of the main factors affecting productivity is the number of plants the nursery can grow per square metre of available space. “The more growing plants you have per square metre, the higher your ultimate production will be. But you can’t place orchids too close together or they will get damaged. The leaves grow out over the side of the pot and the roots also tend to grow out of the pot and take root in nearby pots. This makes the plants very susceptible to damage when you move them. So before we set out the plants in the second growing area, we fit a transparent plastic collar around each pot.”
This protects the plant, but even more importantly, it holds the leaves together. “This way we can put more pots next to each other. Without the collar, we would get around 56 plants per square metre on the mobile benches in the second growing area; with the collar, we get up to about 65.” That’s around 15% more,” Hazeu points out.
The Hazeu brothers are always on the lookout for new technologies and ways to improve working practices in the nursery and boost the quality of the end product. This is important in terms of pricing, reducing losses and optimising productivity. Using special stakes and clips to secure the flower spikes and fitting plastic collars round the pots have proved to be effective measures, but Hazeu is also exploring the potential of process automation. “Because orchids are so fragile, most operations are still done by hand,” he explains.
That applies to sorting, but also to the tasks of fitting the collars and securing the flower spikes. “We have investigated the potential for using cameras for sorting and quality control, for instance, but as yet that’s too complex and expensive for orchids. We automated the supply of pots for sorting and fitting collars using conveyor belts back in 2012. The most logical next step was to automate the process of setting down pots on the mobile benches.”
System manufacturer Javo was commissioned to build the machine. “We make a lot of machines for filling, moving and handling pots and trays, or equipment that meets health and safety requirements,” Sales Manager Peter Rijnders explains. Omron, one of Javo’s regular suppliers, was also involved in developing the robot. The robot for Hazeu Orchids was the first one to be built on the Sysmac platform, enabling it to be developed and programmed as an integrated whole, including the interface, PLCs and servo drives.
The work the robot does is not particularly complicated in itself. Once the pots have been fitted with a collar, they are brought in on a conveyor. The belt consists of a line of individual pot cups, as orchids in pots are quite unstable. The plants are lined up a pre-programmed distance apart in the robot. Once sensors have checked that the row is full, a “fork” with adjustable tines picks up the row of pots and places it on the mobile bench.
There is an additional subtle movement programmed in. The fork gently nudges the row of pots against the previous row and then backs up slightly before setting down the pots. This prevents the pots from knocking against the leaves or roots of plants in the previous row and falling over. This extra bit of programming has reduced the number of fallen pots by more than 99%, which means less damage and no need to manually intervene to pick up the pots. Because the Omron controller is so user-friendly, it was relatively easy to incorporate this extra movement.
A linear drive is used to ensure precise movements of the fork. The arm equipped with the fork is suspended from a gantry and is controlled using industrial servo motors.
The safety program is quick and easy to configure and adjust in the safety controller thanks to the use of standard safety modules. This makes it possible to seamlessly integrate the safety controller into the application and control unit.
There are two robots installed at Hazeu: one on a line with first-rate plants ready to move on to the next stage of the growing process, and another on a line for plants that are too small as yet and need more time to develop. A program has been written for the robots that calculates how many pots are to be placed on the mobile benches and how they are to be spaced, based on pot and plant size. These profiles are stored in a database and can be easily selected on the colour touch screen, which also displays information about the robot’s status and any error messages.
Since the robots can be communicated with remotely, the two suppliers can constantly monitor their status and performance and can tweak the control software if necessary.
So far the robots have not needed much maintenance, however, and they have been operating problem-free since they were introduced in June 2014. Each robot has a capacity of about 700 cycles of the robot arm per hour, but initially this was not fully utilised. “Currently, all the pots are handled by the robots and we have been gradually stepping up the capacity,” Hazeu explains. “Besides the fact that they prevent damage, the robots have reduced our labour costs by around 20%.”
For some time now, Hazeu Orchids has been working with a robot that automates some of their internal transportation. The pots with collars are brought in on a conveyor belt and lined up in rows in the robot a pre-programmed distance apart. A fork fitted with sensors sets them down on the mobile bench. An additional programmed movement prevents the plants from damaging each other.
Text: Theo Snijders. Images: Leo Duijvestijn and Wim van IJzendoorn.
In the European SWEEPER project, efforts are being made to create a sweet pepper harvesting robot. Part of this project is the development of fruit and obstacle recognition based on colour and depth images.
For this purpose, an artificial 3D model of a pepper crop was recently produced based on plant measurements in the greenhouse, including all the distributions of angles and geometry. This digital model can be used to grow random, unique digital plants in a simulated greenhouse environment. Once the digital crop is created, synthetic colour and depth images can be rendered on a supercomputer and used as training material for artificial intelligent learning systems. This enables us to localise parts of the plants such as the fruits, leaves and stems from the robot’s perspective.
The recognised parts of stems can then be used to map out obstacles for the robot to avoid during harvesting. Previous research has shown that the angle of approach during harvesting is very important in achieving a high harvesting percentage.
Leo van den Harg, of the potted rose nursery in Vierpolders that bears his name, bought a Rombomatic ten years ago: a combination of four robots that takes cuttings from roses and subsequently roots them. ‘In the spring, we use it 7 days a week, 24 hours a day.’
What prompted you to buy this robot?
‘We were on the brink of an expansion, which would mean having to take as many as 400,000 cuttings a week for several weeks at a time. People can easily develop neck and shoulder complaints from this. Also: you can explain precisely to people how you want them to take certain cuttings, and root them, but people will always be people, if you know what I mean. If you use a robot you can be assured of a uniform quality.’
How long does it take for the investment of a robot to pay itself back?
‘Thanks to the robot I need fewer workers to take the cuttings and root them. Still, the reduction in personnel costs did not enable me to earn back my investment immediately. However, after a few expansions our investment in the Rombomatic was amply paid back. The Rombomatic also relieves my workers from heavy manual labour: all they have to do is put the cuttings in position. The only equipment they need to do this comfortably is a good chair.’
Does a robot always perform its duties?
‘Absolutely. In spring we use it 7 days a week, 24 hours a day. Three years ago we had four robots replaced. By that time we had rooted 125 million cuttings. That’s 125 million movements! We began to get an increasing number of malfunctions due to wear and tear. Of course, that wasn’t a huge problem: if one robot was out of operation, the other three can continue working. However, the capacity of the Rombomatic then drops to 80%. A big bonus was that the robots were replaced by new ones that work 10% faster.’
Do you use any other robots in addition to the ones that clip and root cuttings?
‘We make use of a packaging robot, but I wouldn’t really call that a genuine robot. That is more a question of automating repetitive movements. The same applies to the robots that pick up and move plants. Besides that, we don’t have any other duties here that could be performed by a robot.’
Text: Tuinbouwteksten.nl/Mario van Vliet. Photo: Mario Bentvelsen.
‘We won’t even begin to consider developing a solution if the pay-back time is longer than three and a half years,’ says Piet Oomen, director and owner of the ISO Group in Gameren. This company was established in 2009 for the development of machines that has robots integrated in them. Since then, the ISO Group has developed fully-automated robots for numerous crops. ‘An integrated robot doesn’t need to be more expensive than a conventional machine.’
What prompted you to start developing robot applications in horticulture?
‘The year 2009 saw a substantial crisis in the horticulture industry. We noticed that at our machine plant as well. One of the horticulture firms near us rooted millions of cuttings every year. One day, we put our heads together and developed an automated rooting tool for plant cuttings. That was the beginning, and it had an enormous impact on our company. We were soon able to incorporate the combination of vision technology and robotics into a wide variety of applications.’
How is your company working on the development or robot applications?
‘We apply robot applications and vision technology to automated tools for clipping and rooting plant cuttings, grading machines, transplanters and grafting equipment. Working in collaboration with our partners in the field of vision technology, we continually aim to raise the bar even higher. This means: training robots to “see” better, in combination with the accurate and failure-free processing of materials. We are now able to process increasingly smaller or more complex cuttings and are attaining ever higher levels of precision in grading. Thanks to our and Wageningen University and Research Centre’s concerted efforts in the field of 3D vision, we are now able to process the vines of potted roses faster than ever before, and monitor the growth process of such plants with greater precision, too.’
What is the average pay-back time for an ‘average’ robot application?
‘An investment in one of our robot applications will earn itself back between two and three and a half years. This is quite short, actually. However, it is more difficult to get financing for long-term investments. This is why we won’t even consider developing a robot application with a pay-back time longer than that. An integrated robot doesn’t need to be more expensive than a conventional machine. The investment in a robot is usually earned back within a short space of time due to the lower price, simple operation and even lower maintenance costs.’
What is your vision of the future for robots in horticulture?
‘We will continue to make improvements in terms of quality and speed. I have high hopes with regard to phenotyping: the ability to predict quality-related results based on external plant traits. This is highly useful not only in grading plants, but in selecting seeds as well. And my own agenda remains: to continue to remain ahead of the competition with cheap labour. One of our customers decided, for instance, to relocate to Africa, where labour was cheaper. He recently returned to the Netherlands and started using automatic rooting equipment for plant cuttings. A robot application not only saves labour, the uniform level of quality it can produce is just as important. A robot will root all cuttings at precisely the same depth, for example. Robots never do anything in a random way.’
Text: Tuinbouwteksten.nl/Mario van Vliet | Photo: Mario Bentvelsen.