Compact robotic gripper with tandem actuation for selective fruit harvesting

TL;DR

The paper addresses selective fruit harvesting in unstructured canopies, where occlusions and delicate fruit require a small, compliant approach and a strong yet bruise-free grip. It presents a compact tandem gripper that attaches with compliant suction cups and then uses telescoping cam-driven fingers to sweep clutter and clamp the fruit, achieving a grasp strength up to $40~\mathrm{N}$ and a high success rate in cluttered and real-field conditions. The authors provide a static model of the finger mechanism, experimental validation of grasp strength across poses, and field validation in a commercial apple orchard, demonstrating robust performance with clusters and leaves. The work offers a practical, field-ready approach to robotic fruit harvesting with potential applicability to other crops and settings.

Abstract

Selective fruit harvesting is a challenging manipulation problem due to occlusions and clutter arising from plant foliage. A harvesting gripper should i) have a small cross-section, to avoid collisions while approaching the fruit; ii) have a soft and compliant grasp to adapt to different fruit geometry and avoid bruising it; and iii) be capable of rigidly holding the fruit tightly enough to counteract detachment forces. Previous work on fruit harvesting has primarily focused on using grippers with a single actuation mode, either suction or fingers. In this paper we present a compact robotic gripper that combines the benefits of both. The gripper first uses an array of compliant suction cups to gently attach to the fruit. After attachment, telescoping cam-driven fingers deploy, sweeping obstacles away before pivoting inwards to provide a secure grip on the fruit for picking. We present and analyze the finger design for both ability to sweep clutter and maintain a tight grasp. Specifically, we use a motorized test bed to measure grasp strength for each actuation mode (suction, fingers, or both). We apply a tensile force at different angles (0°, 15°, 30° and 45°), and vary the point of contact between the fingers and the fruit. We observed that with both modes the grasp strength is approximately 40 N. We use an apple proxy to test the gripper's ability to obtain a grasp in the presence of occluding apples and leaves, achieving a grasp success rate over 96% (with an ideal controller). Finally, we validate our gripper in a commercial apple orchard.

Figures (9)

• Figure 1: From left to right. Approach: The gripper approaches the fruit with a small form factor to reduce collisions. Grasp: An initial compliant grasp with suction cups is followed by a grasp with cam-driven fingers that wedge between neighboring fruit. Pick: The apple is secured in the gripper during the picking motion.
• Figure 2: Gripper render. Left: Suction mode Velasquez2024 with fingers retracted and in-hand perception components labeled. Middle: Gripper with fingers deployed. Top and bottom right: Gripper's main dimensions with fingers retracted.
• Figure 3: Cam-driven finger mechanism. Left: Exploded view of a single finger and cam tracks. The finger's outer pivot pin slides along the outer path of the tracks and the inner pivot pin slides along the inner path. Middle: Side view of the resulting path followed by the telescoping fingers -- the fingers first curve outwards (yellow areas of the paths) to sweep obstacles away from the targeted fruit before moving inwards (dotted red encircled areas) to secure the fruit. Right: Detail view of linkage at the second region with the corresponding geometric configuration of the force transmission model.
• Figure 4: Power transmission with respect to linear distance x travelled by the lead-screw nut. Continuous black curve represents force transmission ratio between the force normal to the finger pad $F_{out}$ and the thrust force at the lead-screw nut$F_{nut}$. We limit x to $59$ mm to reach a maximum force ratio of $1:1$. Dashed green curve represents the angle $\alpha + \theta$. As this angle approaches $90$° the transmission ratio increases exponentially. Dotted blue curve represents the motor torque $T_{motor}$ required to achieve $30$ N $F_{out}$.
• Figure 5: Load cell tests. Top-left: Grasp strength test with different fruit offsets. Bottom-left: Setup used to measure the normal force exerted by an individual finger. Middle: Grasp strength test with the apple's main axis and string aligned, but varying the gripper alignment. Right: Grasp strength test with the apple's main axis aligned with the gripper and the string oriented at $90\deg$, inducing a rotational moment.