Performance Evaluation of a Harvesting Robot for Sweet Pepper

Abstract

This paper evaluates a robot developed for autonomous harvesting of sweet peppers in a commercial greenhouse. Objectives were to assess robot performance under unmodified and simplified crop conditions, using two types of end effectors (Fin Ray; Lip type), and to evaluate the performance contribution of stem‐dependent determination of the grasp pose. We describe and discuss the performance of hardware and software components developed for fruit harvesting in a complex environment that includes lighting variation, occlusions, and densely spaced obstacles. After simplifying the crop, harvest success significantly improved from 6% to 26% (Fin Ray) and from 2% to 33% (Lip type). We observed a decrease in stem damage and an increase in grasp success after enabling stem‐dependent determination of the grasp pose. Generally, the robot had difficulty in successfully picking sweet peppers and we discuss possible causes. The robot's novel capability of perceiving the stem of a plant may serve as useful functionality for future robots.

An autonomous strawberry‐harvesting robot: Design, development, integration, and field evaluation

“…To assess the picking speed of the robot, the picking times for both the one‐arm and dual‐arm modes were calculated from video recordings of the movement. Researchers proposed a definition for cycle harvesting time, which includes perception operation, manipulation of a fruit, placement of the detached fruit, and also the arm traveling time to the next fruit (Bac et al, , ). Due to the variation in robots and crops, similar metrics have been used by other works but with some differences, for example, without counting the time for the arm traveling to the next fruit (Lehnert et al, ) or without adding the perception time (Silwal et al, ).…”

“…An apple robotic harvester was designed and evaluated with an overall success rate of 84% and an average picking time of 6.0 s per fruit; however, they encountered challenges, such as obstacle detection and avoidance (Silwal et al, ). A sweet pepper‐harvesting robot achieved success rates of between 26% and 33% in a modified environment and a cycle time of 94 s for a full harvesting operation (Bac et al, ). Similarly, another sweet pepper‐harvesting robot, named Harvey, achieved a 46% success rate for unmodified crops and 58% for modified crops, with average picking times of 35–40 s (Silwal et al, ).…”

Analysis of a motion planning problem for sweet-pepper harvesting in a dense obstacle environment

“…The constrained-azimuth method avoids challenging planning problems by selecting a different goal configuration. Furthermore, the constrained-azimuth method may result in less damages to the crop because, during robotic harvesting tests trails (Bac et al, 2015), it was noticed that the manipulator sometimes collided with unseen plant parts when it approached a fruit from the side or back side of the stem.…”

“…Preferably, azimuth angle should be selected such that the end-effector is positioned in front of the fruit, with the stem located behind the fruit (Fig. 9), because preliminary field tests of the endeffector mechanism revealed that such a pose increased the probability of successful fruit detachment while reducing the probability of damaging the fruit or stem (Bac et al, 2015). However, accessing this preferred pose is sometimes problematic for fruit located on the left, right, or back side of the stem, for two reasons.…”

A field‐tested robotic harvesting system for iceberg lettuce

“…Many research projects have been performed, but little has filtered through into the commercial world. The more successful projects include a harvester for apples (Silwal et al, ) using a suction method, rice harvesting using custom harvesting systems (Kurita, Iida, Cho, & Suguri, ), and a sweet pepper harvesting system (Bac et al, ). There has also been significant work in the development of autonomous weeding or grading systems including a sugar beet classifying system (Lottes, Hörferlin, Sander, & Stachniss, ) and a grape pruning system (Botterill et al, ).…”