US-20260127325-A1 - MACHINE-LEARNING VIRTUALIZATION-ENABLED HARVESTING

US20260127325A1US 20260127325 A1US20260127325 A1US 20260127325A1US-20260127325-A1

Abstract

A harvesting program system iteratively generates current harvesting programs for performance by harvesting equipment on a mushroom bed. The system receives current mushroom bed data corresponding to the mushroom bed including growing mushrooms at the current times. The system processes the current mushroom bed data using a mushroom bed model to generate current virtual mushroom beds corresponding to current states of the mushroom bed at the current times. The mushroom bed model is trained using labelled training mushroom bed data including known values of the mushroom bed, and using previously-generated virtual mushroom beds corresponding to predicted states of the mushroom bed. The system generates using the mushroom bed model predicted virtual mushroom beds corresponding to predicted states of the mushroom bed at future times. The system generates current harvesting programs based on the predicted virtual mushroom beds, and transmits them performance by the harvesting equipment on the mushroom bed.

Inventors

Peter MANKOWSKI
Vijaya Sankar Velayudham JAYASHREE
Nathan Tomlinson

Assignees

4AG ROBOTICS INC.

Dates

Publication Date: 20260507
Application Date: 20251219

Claims (20)

1 . An automatic harvesting system comprising: at least one processor; and at least one computer-readable medium storing instructions executable by the at least one processor to perform operations comprising: a) controlling harvesting equipment to automatically collect current mushroom bed data corresponding to a mushroom bed including growing mushrooms at a current time; b) processing the current mushroom bed data using a trained mushroom bed model to generate a current virtual mushroom bed corresponding to a current state of the mushroom bed at the current time; c) for each of a plurality of proposed harvesting programs: generating, using the trained mushroom bed model, a corresponding predicted virtual mushroom bed corresponding to a predicted state of the mushroom bed at a future time based on the proposed harvesting program and the current state of the mushroom bed; and calculating a corresponding outcome based on the corresponding predicted virtual mushroom bed; d) selecting as a selected harvesting program the proposed harvesting program corresponding to the outcome best matching predefined optimal outcome parameters; and e) controlling the harvesting equipment to automatically perform the selected harvesting program on the mushroom bed.
2 . The automatic harvesting system of claim 1 , wherein: the predefined optimal outcome parameters comprise a maximum mushroom density.
3 . The automatic harvesting system of claim 2 , wherein: the selected harvesting program comprises pruning a specific mushroom at a particular location in the mushroom bed to reduce a density of mushrooms at the particular location.
4 . The automatic harvesting system of claim 1 , wherein the operations further comprise: f) collecting new current mushroom bed data corresponding to the mushroom bed at the future time; g) processing the new current mushroom bed data using the trained mushroom bed model to generate a new current virtual mushroom bed corresponding to a new current state of the mushroom bed at the future time; h) further training the trained mushroom bed model comprising: comparing the new current virtual mushroom bed with the predicted vitual mushroom bed corresponding to the selected harvesting program to generate differences; and updating parameters of the trained mushroom bed model based on the differences.
5 . The automatic harvesting system of claim 4 , wherein the operations further comprise, after operation h), iteratively repeating operations a) through e).
6 . The automatic harvesting system of claim 4 , wherein the operations comprise iteratively repeating operations a) through h).
7 . The automatic harvesting system of claim 1 , wherein: the harvesting equipment comprises: a harvesting device operable to perform the selected harvesting program on the mushroom bed, wherein operation e) comprises controlling the harvesting device to automatically perform the selected harvesting program on the mushroom bed; and an optical imager operable to collect images of the mushroom bed, wherein the current mushroom bed data comprises the images, wherein operation a) comprises controlling the optical imager to collect the images of the mushroom bed including the growing mushrooms at the current time.
8 . The automatic harvesting system of claim 7 , wherein: the mushroom bed data further comprises respective collection locations of the images, wherein the collection locations indicate corresponding locations on the mushroom bed of a field of view of the optical imager.
9 . The automatic harvesting system of claim 7 , wherein: the current harvesting program comprises a sequence of actions performable by the harvesting device, and operation e) comprises controlling the harvesting device to automatically perform the sequence of actions on the mushroom bed, wherein at least one of the actions comprises harvesting one or more mushrooms at corresponding locations in the mushroom bed.
10 . The automatic harvesting system of claim 7 , wherein: operation a) comprises controlling the optical imager to collect the images at respectively different times while moving the optical imager along a path from a first location in the mushroom bed to a second location in the mushroom bed.
11 . The automatic harvesting system of claim 10 , wherein: operation a) comprises controlling the optical imager to collect the images continuously while the optical imager is in motion along the path from the first location in the mushroom bed to the second location in the mushroom bed.
12 . A computer-implemented method for automatically harvesting mushrooms from a mushroom bed, the method comprising: a) controlling harvesting equipment to automatically collect current mushroom bed data corresponding to a mushroom bed including growing mushrooms at a current time; b) processing the current mushroom bed data using a trained mushroom bed model to generate a current virtual mushroom bed corresponding to a current state of the mushroom bed at the current time; c) for each of a plurality of proposed harvesting programs: generating, using the trained mushroom bed model, a corresponding predicted virtual mushroom bed corresponding to a predicted state of the mushroom bed at a future time based on the proposed harvesting program and the current state of the mushroom bed; and calculating a corresponding outcome based on the corresponding predicted virtual mushroom bed; d) selecting as a selected harvesting program the proposed harvesting program corresponding to the outcome best matching predefined optimal outcome parameters; and e) controlling the harvesting equipment to automatically perform the selected harvesting program on the mushroom bed.
13 . The method of claim 12 , wherein: the predefined optimal outcome parameters comprise a maximum mushroom density.
14 . The method of claim 12 , further comprising: f) collecting new current mushroom bed data corresponding to the mushroom bed at the future time; g) processing the new current mushroom bed data using the trained mushroom bed model to generate a new current virtual mushroom bed corresponding to a new current state of the mushroom bed at the future time; h) further training the trained mushroom bed model comprising: comparing the new current virtual mushroom bed with the predicted vitual mushroom bed corresponding to the selected harvesting program to generate differences; and updating parameters of the trained mushroom bed model based on the differences.
15 . The method of claim 12 , wherein: the harvesting equipment comprises: a harvesting device operable to perform the selected harvesting program on the mushroom bed, wherein operation e) comprises controlling the harvesting device to automatically perform the selected harvesting program on the mushroom bed; and an optical imager operable to collect images of the mushroom bed, wherein the current mushroom bed data comprises the images, wherein operation a) comprises controlling the optical imager to collect the images of the mushroom bed including the growing mushrooms at the current time.
16 . The method of claim 15 , wherein: the mushroom bed data further comprises respective collection locations of the images, wherein the collection locations indicate corresponding locations on the mushroom bed of a field of view of the optical imager.
17 . The method of claim 15 , wherein: the current harvesting program comprises a sequence of actions performable by the harvesting device, and operation e) comprises controlling the harvesting device to automatically perform the sequence of actions on the mushroom bed, wherein at least one of the actions comprises harvesting one or more mushrooms at corresponding locations in the mushroom bed.
18 . The method of claim 15 , wherein: operation a) comprises controlling the optical imager to collect the images at respectively different times while moving the optical imager along a path from a first location in the mushroom bed to a second location in the mushroom bed.
19 . The method of claim 18 , wherein: operation a) comprises controlling the optical imager to collect the images continuously while the optical imager is in motion along the path from the first location in the mushroom bed to the second location in the mushroom bed.
20 . A computer-readable medium storing instructions operable by a processor to perform a method for automatically harvesting mushrooms from a mushroom bed, the method comprising: a) controlling harvesting equipment to automatically collect current mushroom bed data corresponding to a mushroom bed including growing mushrooms at a current time; b) processing the current mushroom bed data using a trained mushroom bed model to generate a current virtual mushroom bed corresponding to a current state of the mushroom bed at the current time; c) for each of a plurality of proposed harvesting programs: generating, using the trained mushroom bed model, a corresponding predicted virtual mushroom bed corresponding to a predicted state of the mushroom bed at a future time based on the proposed harvesting program and the current state of the mushroom bed; and calculating a corresponding outcome based on the corresponding predicted virtual mushroom bed; d) selecting as a selected harvesting program the proposed harvesting program corresponding to the outcome best matching predefined optimal outcome parameters; and e) controlling the harvesting equipment to automatically perform the selected harvesting program on the mushroom bed.

Description

CROSS REFERENCE TO RELATED APPLICATIONS This is a continuation patent application which claims priority under 35 U.S.C. § 120 to U.S. Serial No. 18/929,229, filed October 28, 2024, which claims priority to priority under 35 U.S.C. § 119(e) to provisional patent application U.S. Serial No. 63/594,171, filed October 30, 2023. These applications are hereby incorporated by reference in its entireties herein, including without limitation: the specification, claims, and abstract, as well as any figures, tables, appendices, or drawings thereof. FIELD The present disclosure relates generally to techniques for the cultivation and harvest of agricultural crops, and in particular for the automated cultivation and harvest of mushrooms. BACKGROUND In typical commercial mushroom growing operations, mushrooms are grown in growing beds on the surface of casing soil over substrate in a series of weekly intervals called flushes. Each flush is picked several times per day over a five-day period, and typically two to three flushes are harvested. The size at which the mushrooms are picked depends on market requirements. European and North American commercial production of button mushrooms typically occurs on "Dutch Style" substrate filled shelves, using a two or three flush cropping cycle. The substrate is typically a composted mixture of wheat straw, animal manure, and gypsum. The substrate is pasteurized, inoculated, and colonized with spawn of a selected mushroom strain. The substrate is covered with a casing soil of peat and lime mixture in a layer approximately 45 to 50 mm deep, which is then ruffled with compost added to the casing to mix mushroom mycelium into the casing. Traditionally, commercial mushroom farm operations rely on manual labour to harvest the mushrooms. Manual labour is costly, however, and difficult to optimize. Mushrooms typically grow at such a rate that the mushrooms approximately double in size every 24 hours. Using manual labour, each flush is picked only two or three times per day for the duration of the flush, meaning that a mushroom bed may become overgrown between pickings due to the growth rate of mushrooms. In order to prevent overgrowth of a mushroom bed, a flush can be picked more frequently, but picking at a higher frequency is difficult and costly to accomplish with manual labour. When a bed becomes overgrown, the mushrooms may run out of room and grow into each other, thereby reducing yield, increasing stem growth, and/or causing deformation of each individual mushroom thereby adversely affecting the quality and value of the harvested mushrooms. The automated mushroom harvesting apparatus by Bourdeau et al. disclosed in WIPO International Publication Number WO 2023/010198 A1 solves many of the challenges associated with the automated picking of cultivated mushrooms. There remains, therefore, a need for improved techniques to optimize the total yield and overall effectiveness of automated mushroom cultivation and harvest systems which addresses at least some of the shortcomings of previous solutions and provides yet further advantages, thereby providing a material value over prior techniques. BRIEF DESCRIPTION OF THE DRAWINGS Embodiments will now be described, by way of example only, with reference to the attached Figures. FIG. 1 shows a diagram of an automatic harvesting system. FIG. 2 shows a diagram of a mushroom bed and one embodiment of harvesting equipment. FIG. 3 shows an image of an actual mushroom bed. FIG. 4 shows an image of a synthetic mushroom bed used for training a mushroom bed model. FIG. 5 shows a flowchart of a method for training a mushroom bed model. FIG. 6 shows images of an actual mushroom bed and a corresponding virtual mushroom bed. FIG. 7 shows a flowchart of a method for generating a current harvesting program. It is to be understood that the accompanying drawings are used for illustrating the principles of the embodiments and exemplifications of the subject-matter discussed herein. Hence the drawings are illustrated for simplicity and clarity, and not necessarily drawn to scale and are not intended to be limiting in scope. Reference characters/numbers are used to depict the elements of the subject-matter discussed that are also shown in the drawings. The same corresponding reference characters/numbers are given to a corresponding component or components of the same or similar nature, which may be depicted in multiple drawings for clarity. In particular, specific embodiments or categories of embodiments of an element designated by a particular reference character may be distinguished by means of a suffix, wherein the specific embodiment designated by a reference character having a suffix is a species of the more general element having the same reference character lacking the suffix. For example, an element shown in the drawings and designated by the reference character ###n is a species of the more general element designated by reference character ###, and thus po