Abstract: In some examples, a trap to electrocute a pest may include a logic board. The logic board may create an open circuit using two electrical contacts in touch with two conductive plates in a container located in the trap. A pest entering the container may touch the two conductive plates, causing the circuit to close. Based on determining that the circuit was closed, the logic board may determine that a pest entered the trap. The logic board may cause a shutter inside the trap to move from an approximately horizontal position to an approximately vertical position, thereby causing a portion of a door of the container to close by causing two latch strikes of the door to engage with the two jaws of the container. The circuit may send a wireless notification message, indicating that the pest was captured in the container, over a wireless network to a computing device.

Abstract: In some implementations, a server may receive sensor data from a plurality of detectors, including receiving first sensor data from a first detector and second sensor data from a second detector. The server may use the sensor data to retrain a detector-based machine learning algorithm to create an updated detector-based machine learning algorithm. The server may send the updated detector-based machine learning algorithm to the first detector and the second detector. For example, the updated machine learning algorithm may be trained to identify at least one pest that the machine learning algorithm is incapable of identifying.

Abstract: In some implementations, a pest detector may receive sensor data from a sensor. The pest detector may determine, using a machine learning algorithm, that the sensor data indicates a presence of a first type of pest. The pest detector may send a notification message, including at least a portion of the sensor data, to a computing device and visually indicate that the first type of pest was detected using an indicator. The pest detector may receive an update to the machine learning algorithm from a server and install the update to create an updated machine learning algorithm. The pest detector may receive second sensor data and determine, using the updated machine learning algorithm, that the second sensor data indicates the presence of a second type of pest that is not recognized by the machine learning algorithm.

Abstract: In some implementations, a computing device may receive a notification from a pest detector. In response, the computing device may initiate execution of a software application and display a user interface. The user interface may enable playback of the notification, such as displaying a photograph of a pest or playing back audio of noise made by the pest. The user interface may display a floor plan of a building along with a location of the pest detector and additional locations of additional pest detectors located in the building superimposed on the floor plan. The user interface may enable selecting a particular building of multiple buildings, a particular floor of multiple floors, or a particular room of multiple rooms. The user interface may display where pests have been detected superimposed on the floor plan.

Abstract: In some implementations, a server may receive sensor data from a plurality of detectors, including receiving first sensor data from a first detector and second sensor data from a second detector. The server may use the sensor data to retrain a detector-based machine learning algorithm to create an updated detector-based machine learning algorithm. The server may send the updated detector-based machine learning algorithm to the first detector and the second detector. For example, the updated machine learning algorithm may be trained to identify at least one pest that the machine learning algorithm is incapable of identifying.

Abstract: In some implementations, a pest detector may receive sensor data from a sensor. The pest detector may determine, using a machine learning algorithm, that the sensor data indicates a presence of a first type of pest. The pest detector may send a notification message, including at least a portion of the sensor data, to a computing device and visually indicate that the first type of pest was detected using an indicator. The pest detector may receive an update to the machine learning algorithm from a server and install the update to create an updated machine learning algorithm. The pest detector may receive second sensor data and determine, using the updated machine learning algorithm, that the second sensor data indicates the presence of a second type of pest that is not recognized by the machine learning algorithm.

Abstract: In some implementations, a trap to capture and kill a pest may include a circuit. The circuit may receive sensor data from a sensor and determine that a pest has entered the trap. The circuit may cause a shutter inside the trap to move from an approximately horizontal position to an approximately vertical position, thereby causing a portion of a door of a container located inside the trap to come in contact with adhesive on an outer surface of the container to prevent the pest from leaving the container. Substantially simultaneously with the shutter moving, the circuit may provide electricity to two conductive plates located on in a inner bottom surface of the container, thereby causing the electricity to pass through the pest. The circuit may send a wireless notification message, indicating that the pest was captured in the container, over a wireless network to a computing device.

Abstract: In some implementations, a trap to capture and kill a pest may include a circuit. The circuit may receive sensor data from a sensor and determine that a pest has entered the trap. The circuit may cause a shutter inside the trap to move from an approximately horizontal position to an approximately vertical position, thereby causing a portion of a door of a container located inside the trap to come in contact with adhesive on an outer surface of the container to prevent the pest from leaving the container. Substantially simultaneously with the shutter moving, the circuit may provide electricity to two conductive plates located on in a inner bottom surface of the container, thereby causing the electricity to pass through the pest. The circuit may send a wireless notification message, indicating that the pest was captured in the container, over a wireless network to a computing device.

Abstract: In some examples, a trap to electrocute a pest may include a logic board. The logic board may create an open circuit using two electrical contacts in touch with two conductive plates in a container located in the trap. A pest entering the container may touch the two conductive plates, causing the circuit to close. Based on determining that the circuit was closed, the logic board may determine that a pest entered the trap. The logic board may cause a shutter inside the trap to move from an approximately horizontal position to an approximately vertical position, thereby causing a portion of a door of the container to close by causing two latch strikes of the door to engage with the two jaws of the container. The circuit may send a wireless notification message, indicating that the pest was captured in the container, over a wireless network to a computing device.

Abstract: In some implementations, a detector may include one or more sensors, one or more external indicators, one or more processors, and one or more non-transitory computer readable media storing instructions that are executable by the one or more processors to perform various operations. The operations may include detecting, by a motion sensor, movement associated with the pest and capturing, by a sensor (e.g., an imaging sensor) of the detector, sensor data (e.g., a digital image) of the pest. The operations may include determining, by a machine learning algorithm, that the sensor data indicates a presence of a pest, and sending a notification message to a computing device. The notification message may include at least a portion of the sensor data. The operations may include visually indicating, using a pest indicator of the one or more external indicators, that the pest was detected.

Abstract: In some implementations, a detector may include one or more sensors, one or more external indicators, one or more processors, and one or more non-transitory computer readable media storing instructions that are executable by the one or more processors to perform various operations. The operations may include detecting, by a motion sensor, movement associated with the pest and capturing, by a sensor (e.g., an imaging sensor) of the detector, sensor data (e.g., a digital image) of the pest. The operations may include determining, by a machine learning algorithm, that the sensor data indicates a presence of a pest, and sending a notification message to a computing device. The notification message may include at least a portion of the sensor data. The operations may include visually indicating, using a pest indicator of the one or more external indicators, that the pest was detected.

Abstract: In some implementations, a detector may include one or more sensors, one or more external indicators, one or more processors, and one or more non-transitory computer readable media storing instructions that are executable by the one or more processors to perform various operations. The operations may include detecting, by a motion sensor, movement associated with the pest and capturing, by a sensor (e.g., an imaging sensor) of the detector, sensor data (e.g., a digital image) of the pest. The operations may include determining, by a machine learning algorithm, that the sensor data indicates a presence of a pest, and sending a notification message to a computing device. The notification message may include at least a portion of the sensor data. The operations may include visually indicating, using a pest indicator of the one or more external indicators, that the pest was detected.

Abstract: In some implementations, a detector may include one or more sensors, one or more external indicators, one or more processors, and one or more non-transitory computer readable media storing instructions that are executable by the one or more processors to perform various operations. The operations may include detecting, by a motion sensor, movement associated with the pest and capturing, by a sensor (e.g., an imaging sensor) of the detector, sensor data (e.g., a digital image) of the pest. The operations may include determining, by a machine learning algorithm, that the sensor data indicates a presence of a pest, and sending a notification message to a computing device. The notification message may include at least a portion of the sensor data. The operations may include visually indicating, using a pest indicator of the one or more external indicators, that the pest was detected.

Abstract: In some implementations, a detector may include one or more sensors, one or more external indicators, one or more processors, and one or more non-transitory computer readable media storing instructions that are executable by the one or more processors to perform various operations. The operations may include detecting, by a motion sensor, movement associated with the pest and capturing, by a sensor (e.g., an imaging sensor) of the detector, sensor data (e.g., a digital image) of the pest. The operations may include determining, by a machine learning algorithm, that the sensor data indicates a presence of a pest, and sending a notification message to a computing device. The notification message may include at least a portion of the sensor data. The operations may include visually indicating, using a pest indicator of the one or more external indicators, that the pest was detected.