Abstract: A refuse vehicle includes multiple systems, each system including a sensor. The refuse vehicle also includes an automated check system. The automated check system includes processing circuitry configured to obtain sensor data from the sensor of each of the multiple systems, determine which of the multiple systems require manual inspection based on the sensor data, and operate a display screen to prompt a technician to manually inspect one or more of the multiple systems.

Abstract: A system for detecting and engaging a refuse can includes at least one sensor positioned on a refuse collection vehicle and configured to detect objects on one or more sides of the refuse vehicle, an actuator assembly configured to actuate to engage the refuse can, and a controller configured to detect, using a single-stage object detector, the presence of the refuse can based on first data received from the at least one sensor, determine, based on the first data, a position of the refuse can with respect to the refuse collection vehicle, generate a first trajectory from the refuse collection vehicle to the position of the refuse can, generate a second trajectory for the actuator assembly, and initiate a control action to move at least one of the refuse collection vehicle along the first trajectory or the actuator assembly along the second trajectory to engage the refuse can.

Abstract: A refuse vehicle includes a chassis, an energy storage device supported by the chassis and configured to provide electrical power to a prime mover, wherein activation of the prime mover selectively drives the refuse vehicle, a body for storing refuse therein supported by the chassis, a first hydraulic pump configured to convert electrical power into hydraulic power, a second first hydraulic pump configured to convert electrical power into hydraulic power, and a motor coupled to at least one of the body or the chassis and configured to drive the first hydraulic pump and drive the second hydraulic pump.

Abstract: A refuse vehicle includes multiple systems, each system including a sensor. The refuse vehicle also includes an automated check system. The automated check system includes processing circuitry configured to obtain sensor data from the sensor of each of the multiple systems, determine which of the multiple systems require manual inspection based on the sensor data, and operate a display screen to prompt a technician to manually inspect one or more of the multiple systems.

Abstract: Systems, devices, methods, and computer-readable media for horizon-based navigation. A method can include receiving image data corresponding to a geographical region in a field of view of an imaging unit and in which the device is situated, based on the received image data, generating, by the processing unit, an image horizon corresponding to a horizon of the geographical region and from a perspective of the imaging unit, projecting three-dimensional (3D) points of a 3D point set of the geographical region to an image space of the received image data resulting in a synthetic image, generating, by the processing unit, a synthetic image horizon of the synthetic image, and responsive to determining the image horizon sufficiently correlates with the synthetic image horizon, providing a location corresponding to a perspective of the synthetic image as a location of the processing unit.

Abstract: Systems, devices, methods, and computer-readable media for horizon-based navigation. A method can include receiving image data corresponding to a geographical region in a field of view of an imaging unit and in which the device is situated, based on the received image data, generating, by the processing unit, an image horizon corresponding to a horizon of the geographical region and from a perspective of the imaging unit, projecting three-dimensional (3D) points of a 3D point set of the geographical region to an image space of the received image data resulting in a synthetic image, generating, by the processing unit, a synthetic image horizon of the synthetic image, and responsive to determining the image horizon sufficiently correlates with the synthetic image horizon, providing a location corresponding to a perspective of the synthetic image as a location of the processing unit.

Abstract: A machine receives a large image having large image dimensions that exceed memory threshold dimensions. The large image includes metadata. The machine adjusts an orientation and a scaling of the large image based on the metadata. The machine divides the large image into a plurality of image tiles, each image tile having tile dimensions smaller than or equal to the memory threshold dimensions. The machine provides the plurality of image tiles to an artificial neural network. The machine identifies, using the artificial neural network, at least a portion of the target in at least one image tile. The machine identifies the target in the large image based on at least the portion of the target being identified in at least one image tile.

Abstract: A machine receives, for a given target in a given image, a plurality of annotations, each annotation indicating a range of pixels corresponding to the given target. The machine computes a number of pixels in an intersection of the plurality of annotations. The machine computes an average pixels on target (APOT) for the given target by computing, for each annotation, a number of pixels that corresponds to the given target and taking a mean of those numbers. The machine computes a precision value of the plurality of annotations as the number of pixels in the intersection divided by the APOT. The machine provides an output representing the precision value.

Abstract: A machine receives a large image having large image dimensions that exceed memory threshold dimensions. The large image includes metadata. The machine adjusts an orientation and a scaling of the large image based on the metadata. The machine divides the large image into a plurality of image tiles, each image tile having tile dimensions smaller than or equal to the memory threshold dimensions. The machine provides the plurality of image tiles to an artificial neural network. The machine identifies, using the artificial neural network, at least a portion of the target in at least one image tile. The machine identifies the target in the large image based on at least the portion of the target being identified in at least one image tile.

Abstract: A machine receives, for a given target in a given image, a plurality of annotations, each annotation indicating a range of pixels corresponding to the given target. The machine computes a number of pixels in an intersection of the plurality of annotations. The machine computes an average pixels on target (APOT) for the given target by computing, for each annotation, a number of pixels that corresponds to the given target and taking a mean of those numbers. The machine computes a precision value of the plurality of annotations as the number of pixels in the intersection divided by the APOT. The machine provides an output representing the precision value.