Abstract: An inference system for monitoring a cementitious mixture for three-dimensional printing is provided. The inference system includes an ambient condition sensor, a temperature sensor, a moisture sensor and an image capturing device. The inference system also includes a controller coupled to the ambient condition sensor, the temperature sensor, the moisture sensor, and the image capturing device. The controller receives sensed ambient conditions, a temperature signal, and a moisture content signal. The controller receives an image feed of a portion of a cementitious mixture. The controller also receives signals indicative of a motor speed and a motor torque associated with a mixing container. The controller builds a model and determines a material suitability of the cementitious mixture using the model based on the received ambient conditions, the temperature signal, the moisture content signal, the image feed, the motor speed, and the motor torque and determines one or more corrective actions.

Abstract: A machine control system can store model weights determined via machine learning using a training dataset correlating preset hydraulic valve displacements to measured movement parameters of a machine component. The machine control system can receive an input command for the component and machine state data from machine sensors. A control mapping model can use the model weights to map a combination of the input command and the machine state data into a predicted displacement of the hydraulic valve that causes movement of the component in response to the input command.

Abstract: A system for determining an origin of a fluid from a machine may include a spectral measurement device configured to generate spectral data for a fluid, and a transmitter in communication with the spectral measurement device and configured to transmit the spectral data. The system may also include a receiver in communication with the transmitter and configured to receive a transmission indicative of the spectral data from the transmitter. The system may further include a processor in communication with the receiver and configured to cause execution of an analytical model configured to determine, based at least in part on the spectral data, fluidic information related to the fluid, the fluidic information including an indication of an origin of the fluid. The system may also include an output device in communication with the processor and configured to output the indication of the origin of the fluid.

Abstract: An inference system for monitoring a cementitious mixture for three-dimensional printing is provided. The inference system includes an ambient condition sensor, a temperature sensor, a moisture sensor and an image capturing device. The inference system also includes a controller coupled to the ambient condition sensor, the temperature sensor, the moisture sensor, and the image capturing device. The controller receives sensed ambient conditions, a temperature signal, and a moisture content signal. The controller receives an image feed of a portion of a cementitious mixture. The controller also receives signals indicative of a motor speed and a motor torque associated with a mixing container. The controller builds a model and determines a material suitability of the cementitious mixture using the model based on the received ambient conditions, the temperature signal, the moisture content signal, the image feed, the motor speed, and the motor torque and determines one or more corrective actions.

Abstract: A machine includes a frame, an engine supported on the frame, ground-engaging elements supported on the frame and driven by the engine to propel the machine, a machine light structure supported on the frame, a multiple-axis accelerometer positioned to generate an acceleration signal relative to a base of the machine light structure, and a controller. The controller is programmed to receive the acceleration signal, partition the acceleration signal into overlapped window periods, transform the acceleration signal of each overlapped window period into a power spectral density, and pass the power spectral densities of each overlapped window period through a strain power spectral densities prediction model to generate predicted strain power spectral densities of the machine light structure. A health monitor receives the predicted strain power spectral densities and is configured to generate a notification responsive to the predicted strain power spectral densities.

Abstract: A machine control system can store model weights determined via machine learning using a training dataset correlating preset hydraulic valve displacements to measured movement parameters of a machine component. The machine control system can receive an input command for the component and machine state data from machine sensors. A control mapping model can use the model weights to map a combination of the input command and the machine state data into a predicted displacement of the hydraulic valve that causes movement of the component in response to the input command.

Abstract: A method of using a virtual machine operator model includes providing a virtual machine operating environment having a current state and including a virtual machine and a virtual operator agent acting within the virtual machine operating environment. The method also includes executing a control action, by the virtual operator agent, relative to the virtual machine based on the virtual machine operator model. The method also includes analyzing a new state of the virtual machine operating environment resulting from execution of the control action, assigning a positive reward or a negative reward to the control action based on the new state, assigning the negative reward to the control action resulting in a high implement jerk movement of the virtual machine, and executing a learning algorithm to update the virtual machine operator model based on the positive reward or the negative reward.

Abstract: A system for determining an origin of a fluid from a machine may include a spectral measurement device configured to generate spectral data for a fluid, and a transmitter in communication with the spectral measurement device and configured to transmit the spectral data. The system may also include a receiver in communication with the transmitter and configured to receive a transmission indicative of the spectral data from the transmitter. The system may further include a processor in communication with the receiver and configured to cause execution of an analytical model configured to determine, based at least in part on the spectral data, fluidic information related to the fluid, the fluidic information including an indication of an origin of the fluid. The system may also include an output device in communication with the processor and configured to output the indication of the origin of the fluid.

Abstract: A method of using a virtual machine operator model includes providing a virtual machine operating environment having a current state and including a virtual machine and a virtual operator agent acting within the virtual machine operating environment. The method also includes executing a control action, by the virtual operator agent, relative to the virtual machine based on the virtual machine operator model. The method also includes analyzing a new state of the virtual machine operating environment resulting from execution of the control action, assigning a positive reward or a negative reward to the control action based on the new state, assigning the negative reward to the control action resulting in a high implement jerk movement of the virtual machine, and executing a learning algorithm to update the virtual machine operator model based on the positive reward or the negative reward.

Abstract: A machine includes a frame, an engine supported on the frame, ground-engaging elements supported on the frame and driven by the engine to propel the machine, a machine light structure supported on the frame, a multiple-axis accelerometer positioned to generate an acceleration signal relative to a base of the machine light structure, and a controller. The controller is programmed to receive the acceleration signal, partition the acceleration signal into overlapped window periods, transform the acceleration signal of each overlapped window period into a power spectral density, and pass the power spectral densities of each overlapped window period through a strain power spectral densities prediction model to generate predicted strain power spectral densities of the machine light structure. A health monitor receives the predicted strain power spectral densities and is configured to generate a notification responsive to the predicted strain power spectral densities.