More and more harvesting robots are being used worldwide – especially for high-priced fruit – primarily to replace missing workers. For example, robots pick strawberries, tomatoes, apples, peppers or kiwi.
Advertisement
Basically, there are two different approaches: the robot shakes the tree or bush and then collects the falling fruit, or – especially in the case of sensitive fruit – the machine approaches the object to be harvested with a gripper and picks it. However, some fruits, such as raspberries, are extremely delicate – and are therefore easily destroyed by robotic grippers.
Training on artificial raspberries
Kai Junge from the EPFL in Lausanne, together with colleagues, has now trained a robot to solve this problem with the help of a “physical twin” in the form of an artificial raspberry equipped with sensors.
Recommended Editorial Content
With your consent, an external YouTube video (Google Ireland Limited) will be loaded here.
Always load YouTube video Load YouTube video now
The artificial raspberry consists of a 3D printed outer skin that is filled with liquid silicone. The external pressure force when gripping the fruit is measured by a fluidic sensor. The sensor raspberry is attached with a magnet to a flexible, artificial tendril, which is also 3D printed. The distance between the two halves of the magnet defines how great the maximum pulling force is to detach the sensory raspberry from the tendril.
Junge and colleagues first had people pick this sensory raspberry, changing the parameters and recording the force profiles. They then used it to optimize the parameters for controlling the gripper. The robotic gripper, which sits on a standard robotic arm, has a Raspi camera that first locates and approaches a raspberry. Then two soft silicone fingers close around the berry and the robot starts to pull until the fruit comes loose. As soon as this happens, the controller should ensure that the lateral force on the berry is relaxed as much as possible – without letting go of the fruit. The authors explain technical details in their paper.
Advertisement
Robot was having trouble finding raspberries
With the training on the physical twin, the authors primarily want to make expensive and lengthy field tests superfluous. Because if such a system works in the laboratory, it is still a long way from being able to use it in practice. In fact, the hardware and software first have to be adapted to the complex conditions in use. However, this is usually only possible if there are also fruits ready to be harvested for training – this period is limited, however.
With the model raspberry, the researchers therefore ran through a whole range of parameters – rigidity, force to detach, etc. – in order to make the controller as universally applicable as possible. Nevertheless, the practical test showed that the machine correctly harvested 80 percent of the ripe fruit. However, the robot had problems finding raspberries hidden by tendrils, leaves or fruits and in some cases could not muster enough strength.
A similar project shows how lengthy the step from the laboratory to practice can be: The start-up Fieldwork Robotics, which was founded by the agricultural robotics group at the University of Plymouth, has had two copies of a raspberry Harvesting robots in practical use.
According to the Guardian, the four-armed robots could pick a kilo of fruit an hour at that point, which should increase to four kilograms an hour by 2023. The first version of the robots, which have been in development since 2018, took about a minute per berry.
(wst)
To home page