Abstract: Sorting biogenic material from a stream of heterogeneous materials is disclosed, including: detecting biogenic material within an input stream of heterogeneous material; sorting the biogenic material based at least in part on a desired biochar formulation; and tracking a composition of a sorted mixture of biogenic material.

Abstract: Obtaining biogenic material from a stream of heterogeneous materials is disclosed, including: receiving an input stream of heterogeneous material; separating a sub-stream of at least biogenic material from the input stream of heterogeneous material using a screen; removing a set of non-biogenic material from the sub-stream of at least biogenic material based at least in part on density separation; and drying the sub-stream of at least biogenic material after removal of the set of non-biogenic material, wherein the sub-stream of at least biogenic material after removal of the set of non-biogenic material comprises biogenic material that is suitable to produce biochar.

Abstract: Using machine learning to recognize variant objects is disclosed, including: identifying an object as a variant of an object type by inputting sensed data associated with the object into a modified machine learning model corresponding to the variant of the object type, wherein the modified machine learning model corresponding to the variant of the object type is generated using a machine learning model corresponding to the object type; and generating a control signal to provide to a sorting device that is configured to perform a sorting operation on the object, wherein the sorting operation on the object is determined based at least in part on the variant of the object type associated with the object.

Abstract: Separation of materials using sorting devices is disclosed, including: receiving image data from sensors, wherein the image data corresponds to a set of objects being transported through a base sorting line; determining a subset of objects from the set of objects that are associated with a designated category based on the image data; sending a first set of data to a first sorting device to perform a diverting action on the subset of objects, wherein the first sorting device and a second sorting device are located along the base sorting line, wherein each of the first and second sorting devices is configured to perform diverting actions to separate materials associated with the designated category off from the base sorting line; and sending a second set of data to the second sorting device to perform a second action, wherein the second action is determined based on the image data.

Abstract: Using machine learning to recognize variant objects is disclosed, including: identifying an object as a variant of an object type by inputting sensed data associated with the object into a modified machine learning model corresponding to the variant of the object type, wherein the modified machine learning model corresponding to the variant of the object type is generated using a machine learning model corresponding to the object type; and generating a control signal to provide to a sorting device that is configured to perform a sorting operation on the object, wherein the sorting operation on the object is determined based at least in part on the variant of the object type associated with the object.

Abstract: A reconfigurable sorting facility is disclosed, including: a plurality of sorting facility devices; and one or more processors configured to: store physical layout information that describes corresponding positions of the plurality of sorting facility devices within a sorting facility; collect feedback data from the plurality of sorting facility devices; and identify at least a subset of the plurality of sorting facility devices to reconfigure based at least in part on the collected feedback data and respective positions of the subset of the plurality of sorting facility devices.

Abstract: Time-varying control of a controllable air stream is disclosed, including: a set of image sensors configured to capture image data showing a target object being conveyed by a conveyor device along a Y-direction; an array of air jets located along an X-direction, wherein the X-direction crosses the Y-direction; and a processor configured to: receive the image data from the set of image sensors; determine a classification of the target object within the image data; and modify a two-dimensional firing polygon of air to be directed by the array of air jets on the target object based at least in part on the classification of the target object; and wherein the array of air jets is configured to perform a sorting operation on the target object as the target object is translated across the array of air jets based at least in part on the modified two-dimensional firing polygon.

Abstract: Co-located sensors in a sorting facility are disclosed, including: a first sensor associated with a first field of view, and wherein the first sensor is located in a sorting facility; a second sensor associated with a second field of view, wherein the second field of view is adjacent to or overlaps with the first field of view, and wherein the second sensor is located in the sorting facility; and one or more processors configured to: receive a first image including a first set of objects from the first sensor; receive a second image including a second set of objects from the second sensor; and evaluate the first image and the second image to determine a sorting decision with respect to a target object.

Abstract: Centralized control of multiple sorting facilities is disclosed, including: a feedback interface configured to collect sorting results feedback data from a first sorting facility, wherein the first sorting facility is associated with a second sorting facility; and one or more processors configured to reconfigure a set of sorting facility devices located in the second sorting facility based at least in part on the sorting results feedback data.

Abstract: Dynamic bale creation is disclosed, including: a plurality of sorting facility devices configured to capture objects from a material stream based at least in part on target sorting metrics specified by a bale recipe; and one or more processors configured to: determine current sorting metrics associated with captured objects that are to be added to a collection mechanism at a sorting facility; and determine whether to trigger reconfiguration of at least a subset of the plurality of sorting facility devices based at least in part on a comparison between the target sorting metrics and the current sorting metrics.

Abstract: Heterogenous material sorting is disclosed, including: identifying a first target object associated with a first object type on a surface based at least in part on a first sensed signal; providing a first control signal to a first sorting device to cause the first sorting device to remove the first target object from the surface, wherein the first sorting device is configured to manipulate objects associated with the first object type; identifying a second target object associated with a second object type on the surface based at least in part on a second sensed signal; and providing a second control signal to a second sorting device to cause the second sorting device to remove the second target object from the surface, wherein the second sorting device is configured to manipulate objects associated with the second object type.

Abstract: Using machine learning to recognize variant objects is disclosed, including: identifying an object as a variant of an object type by inputting sensed data associated with the object into a modified machine learning model corresponding to the variant of the object type, wherein the modified machine learning model corresponding to the variant of the object type is generated using a machine learning model corresponding to the object type; and generating a control signal to provide to a sorting device that is configured to perform a sorting operation on the object, wherein the sorting operation on the object is determined based at least in part on the variant of the object type associated with the object.

Abstract: Object material type identification using multiple types of sensors is disclosed, including: obtaining a machine learning model, wherein the machine learning model has been trained using training data comprising vision sensor data on a set of objects, and wherein the vision sensor data on the set of objects is associated with material characteristic labels that are determined based at least in part on non-vision sensor data on the set of objects; obtaining a vision sensor signal corresponding to an object; and using the machine learning model and the vision sensor signal to determine a material characteristic type associated with the object.

Abstract: Cloud and facility-based machine learning for sorting facilities is disclosed, including: obtaining a machine learning model associated with a domain associated with materials to be sorted at a first sorting facility; and generating a modified machine learning model by training the machine learning model using data obtained from the first sorting facility.

Abstract: Maintaining a data structure corresponding to a target object is disclosed, including: determining that an identified target object from a sensed signal is a new target object, wherein the sensed signal is generated at a sorting facility; generating a new data structure corresponding to the new target object; and updating the new data structure with an attribute associated with the new target object, wherein the attribute is derived from the sensed signal associated with the new target object.