Built Robotics Inc.

An autonomous earth moving system can determine a desired state for a portion of the EMV including at least one control surface. Then the EMV selects a set of control signals for moving the portion of the EMV from the current state to the desired state using a machine learning model trained to generate control signals for moving the portion of the EMV to the desired state based on the current state. After the EMV executes the selected set of control signals, the system measures an updated state of the portion of the EMV. In some cases, this updated state of the EMV is used to iteratively update the machine learning model using an online learning process.

An excavation vehicle capable of autonomously actuating an excavation tool or navigating an excavation vehicle to perform an excavation routine within an excavation site is described herein. Sensors mounted to the excavation vehicle and the excavation tool produce signals representative of a position and orientation of the corresponding joint relative on the excavation vehicle relative to the excavation site, a position and orientation of the excavation vehicle relative to the excavation site, and one or more features of the excavation site based on the position of the excavation vehicle within the excavation site. A set of solenoids are configured to couple to corresponding hydraulic valves of the excavation tool to actuate the valve. A controller produces actuating signals to control the joints of the excavation tool to autonomously perform the excavation routine based on the signals produced by the sensors.

This description provides an autonomous or semi-autonomous excavation vehicle that is capable of navigating through a dig site and carrying an excavation routine using a system of sensors physically mounted to the excavation vehicle. The sensors collect one or more of spatial, imaging, measurement, and location data representing the status of the excavation vehicle and its surrounding environment. Based on the collected data, the excavation vehicle executes instructions to perform an excavation routine by excavating earth from a hole using an excavation tool positioned at a single location within the site. The excavation vehicle is also able to carry out numerous other tasks, such as checking the volume of excavated earth in an excavation tool, navigating the excavation vehicle over a distance while continuously excavating earth from a below surface depth, and preparing a digital terrain model of the site as part of a process for creating the excavation routine.

An autonomous earth moving system can determine a desired state for a portion of the EMV including at least one control surface. Then the EMV selects a set of control signals for moving the portion of the EMV from the current state to the desired state using a machine learning model trained to generate control signals for moving the portion of the EMV to the desired state based on the current state. After the EMV executes the selected set of control signals, the system measures an updated state of the portion of the EMV. In some cases, this updated state of the EMV is used to iteratively update the machine learning model using an online learning process.

When an EMV performs an action comprising moving a tool of the EMV through soil or other material, the EMV can measure a current speed of the tool through the material and a current kinematic pressure exerted on the tool by the material. Using the measured current speed and kinematic pressure, the EMV system can use a machine learned model to determine one or more soil parameters of the material. The EMV can then make decisions based on the soil parameters, such as by selecting a tool speed for the EMV based on the determined soil parameters.

In some implementations, the EMV uses a calibration to inform autonomous control over the EMV. To calibrate an EMV, the system first selects a calibration action comprising a control signal for actuating a control surface of the EMV. Then, using a calibration model comprising a machine learning model trained based on one or more previous calibration actions taken by the EMV, the system predicts a response of the control surface to the control signal of the calibration action. After the EMV executes the control signal to perform the calibration action, the EMV system monitors the actual response of the control signal and uses that to update the calibration model based on a comparison between the predicted and monitored states of the control surface.

An earth moving vehicle (EMV) autonomously performs an earth moving operation within a dig site. If the EMV determines that a state of the EMV or the dig site triggers a triggering condition associated with a pause in the autonomous behavior of the EMV, the EMV determines a risk associated with the state or triggering condition. If the risk is greater than a first threshold, the EMV continues the autonomous performance and notifies a remote operator that the triggering condition was triggered. If the risk is greater than the first threshold risk but less than a second threshold risk, the EMV is configured to operate in a default state before continuing and notifying the remote operator. If the risk is greater than the second threshold risk, the EMV notifies the remote operator of the state pauses the performance until feedback is received from the remote operator.

This description provides an autonomous or semi-autonomous excavation vehicle that is capable of navigating through a dig site and carrying out an excavation routine using a system of sensors physically mounted to the excavation vehicle. The sensors collects any one or more of spatial, imaging, measurement, and location data representing the status of the excavation vehicle and its surrounding environment. Based on the collected data, the excavation vehicle executes instructions to carry out an excavation routine. The excavation vehicle is also able to carry out numerous other tasks, such as checking the volume of excavated earth in an excavation tool, and helping prepare a digital terrain model of the site as part of a process for creating the excavation routine.

An autonomous earth moving system can determine a desired state for a portion of the EMV including at least one control surface. Then the EMV selects a set of control signals for moving the portion of the EMV from the current state to the desired state using a machine learning model trained to generate control signals for moving the portion of the EMV to the desired state based on the current state. After the EMV executes the selected set of control signals, the system measures an updated state of the portion of the EMV. In some cases, this updated state of the EMV is used to iteratively update the machine learning model using an online learning process.

When an EMV performs an action comprising moving a tool of the EMV through soil or other material, the EMV can measure a current speed of the tool through the material and a current kinematic pressure exerted on the tool by the material. Using the measured current speed and kinematic pressure, the EMV system can use a machine learned model to determine one or more soil parameters of the material. The EMV can then make decisions based on the soil parameters, such as by selecting a tool speed for the EMV based on the determined soil parameters.