Lab2Field transfer of a robotic raspberry harvester enabled by a soft sensorized physical twin

Abstract

Robotic fruit harvesting requires dexterity to handle delicate crops and development relying upon field testing possible only during the harvesting season. Here we focus on raspberry crops, and explore how the research methodology of harvesting robots can be accelerated through soft robotic technologies. We propose and demonstrate a physical twin of the harvesting environment: a sensorized physical simulator of a raspberry plant with tunable properties, used to train a robotic harvester in the laboratory regardless of season. The sensors on the twin allow for direct comparison with human demonstrations, used to tune the robot controllers. In early field demonstrations, an 80% harvesting success rate was achieved without any modifications on the lab trained robot.

Soft robotics for farm to fork: applications in agriculture & farming

“…In addition to end-effectors developed for specific crops, some consideration has also been given to multi-purpose soft grippers such as [45] and [46], the latter being re-sizable depending on the targeted fruit or vegetable. In the context of raspberry harvesting, there has also been work on developing soft robotic simulators to train robots for harvesting [47], to enable rapid deployment to the field for such delicate crops [13]. Three diverse examples of harvesting solutions spanning from raspberries to peppers are summarized in figure 4.…”

“…Here, soft robotic technologies include sensors, actuators, and compliant structures which enable the development of robots that exploit the inherent softness in their bodies for increasingly complex interactions with the environment [12]. These technologies, combined with advancements in soft robotics modeling and [13]) Mark Stebnicki / Pexels.com. Quang Nguyen Vinh / Pexels.com.…”

“…Quang Nguyen Vinh / Pexels.com. Reproduced from [13]. CC BY 4.0. optimization methods, can lead to more capable robots that can aid productivity, remove the need for humans to perform labor-intensive tasks, and contribute to making agriculture more sustainable.…”

Editors’ Choice 2023

“…However, selecting ripe fruit and picking without damaging the product is a challenging task for a robot. Kai Junge and colleagues reported in Communications Engineering 13 an artificial plant complete with fake raspberries designed to allow robots to be trained to pick fruit throughout the year, and not just during summer when ripe fruit is available for field trials (see Fig. 2 ).…”