Abstract: Examples of systems and methods for locating movable objects such as carts (e.g., shopping carts) are disclosed. Such systems and methods can use dead reckoning techniques to estimate the current position of the movable object. Various techniques for improving accuracy of position estimates are disclosed, including compensation for various error sources involving the use of magnetometer and accelerometer, and using vibration analysis to derive wheel rotation rates. Also disclosed are various techniques to utilize characteristics of the operating environment in conjunction with or in lieu of dead reckoning techniques, including characteristic of environment such as ground texture, availability of signals from radio frequency (RF) transmitters including precision fix sources. Such systems and methods can be applied in both indoor and outdoor settings and in retail or warehouse settings.

Abstract: A system for monitoring shopping baskets (e.g., baskets on human-propelled carts, motorized carts, or hand-carried baskets) can include a computer vision unit that can image a surveillance region (e.g., an exit to a store), determine whether a basket is empty or loaded with merchandise, and assess a potential for theft of the merchandise. The computer vision unit can include a camera and an image processor programmed to execute a computer vision algorithm to identify shopping baskets and determine a load status of the basket. The computer vision algorithm can comprise a neural network. The system can identify an at least partially loaded shopping basket that is exiting the store, without indicia of having paid for the merchandise, and execute an anti-theft action, e.g., actuating an alarm, notifying store personnel, activating a store surveillance system, activating an anti-theft device associated with the basket (e.g., a locking shopping cart wheel), etc.

Abstract: Examples of systems and methods for calibrating or operating a magnetic sensor for sensor temperature or operating conditions are provided. The magnetic sensor can comprise a dual magnetometer sensor that comprises a first, low-power-consumption magnetometer (e.g., a magneto-inductive magnetometer) and a second higher-power-consumption magnetometer (e.g., a magneto-resistive magnetometer). The second magnetometer can have a lower unit-to-unit variation in temperature calibration parameters and can be used to temperature-correct readings from the first magnetometer. The magnetic sensor can dynamically switch between usage of the first magnetometer and the second magnetometer in order to provide a dynamic sample rate that can depend on conditions within the sensor or external to the sensor.

Abstract: A wheel or wheel assembly for a non-motorized vehicle, such as a shopping cart, is disclosed that detects its direction of rotation. In one embodiment, the wheel assembly includes a plurality of magnets mounted to a rotating portion of the wheel, and includes a magnetic sensor, such as a tunneling magnetoresistance sensor, mounted to a non-rotating portion. As the wheel rotates the magnets produce a varying magnetic field that is sensed by the sensor, which outputs a signal corresponding to the sensed magnetic field. The magnets are arranged—preferably asymmetrically—such that the sensor's output signal differs depending upon whether the wheel is rotating in the clockwise versus counterclockwise direction. A controller analyzes the sensor's output signal to determine the direction of rotation. In another embodiment, the magnets are replaced by conductive targets, and an eddy current sensor is used for the magnetic sensor.

Abstract: A system for monitoring shopping baskets (e.g., baskets on human-propelled carts, motorized carts, or hand-carried baskets) can include a computer vision unit that can image a surveillance region (e.g., an exit to a store), determine whether a basket is empty or loaded with merchandise, and assess a potential for theft of the merchandise. The computer vision unit can include a camera and an image processor programmed to execute a computer vision algorithm to identify shopping baskets and determine a load status of the basket. The computer vision algorithm can comprise a neural network. The system can identify an at least partially loaded shopping basket that is exiting the store, without indicia of having paid for the merchandise, and execute an anti-theft action, e.g., actuating an alarm, notifying store personnel, activating a store surveillance system, activating an anti-theft device associated with the basket (e.g., a locking shopping cart wheel), etc.

Abstract: A system for monitoring shopping carts uses cameras to generate images of the carts moving in a store. In some implementations, cameras may additionally or alternatively be mounted to the shopping carts and configured to image cart contents. The system may use the collected image data, and/or other types of sensor data (such as the store location at which an item was added to the basket), to classify items detected in the shopping carts. For example, a trained machine learning model may classify item in a cart as “non-merchandise,” “high theft risk merchandise,” “electronics merchandise,” etc. When a shopping cart approaches a store exit without any indication of an associated payment transaction, the system may use the associated item classification data, optionally in combination with other data such as cart path data, to determine whether to execute an anti-theft action, such as locking a cart wheel or activating a store alarm.

Abstract: A system for monitoring and controlling shopping cart usage comprises a wheel assembly that attaches to a shopping cart. In some embodiments the wheel assembly includes a wheel, a brake that can be activated to inhibit rotation of the wheel, a controller that controls the brake, a VLF receiver, and an RF transceiver. The RF transceiver may, for example, operate in a 2.4 GHz frequency band. In some implementations, the RF transceiver may be used to detect entry of the shopping cart into a checkout area of the store, and the VLF receiver may be used to detect that the shopping cart is exiting the store. The controller may activate the brake if the shopping cart attempts to exit the store without first passing through a checkout area.

Abstract: A self-leveling caster assembly includes a caster frame and a castor fork which is configured to pivot relative to the caster frame. A spring or other biasing mechanism biases the caster fork away from the caster frame, to maintain contact between the ground and a wheel supported by the caster fork along a travel range of the wheel. Even if a frame supporting the self-leveling caster assembly is bent, pulling the wheel away from the ground, the biased caster fork will maintain contact between the ground and the wheel, allowing continued use of a shopping cart or other wheeled apparatus including the self-leveling caster assembly.

Abstract: A navigation system uses a dead reckoning method to estimate an object's present position relative to one or more prior positions. In some embodiments, the dead reckoning method determines a change in position from the object's heading and speed during an elapsed time interval. In embodiments suitable for use with wheeled objects, the dead reckoning method determines the change in position by measuring the heading and the amount of wheel rotation. Some or all of the components of the navigation system may be disposed within a wheel, such as a wheel of a shopping cart.

Abstract: Examples of systems and methods for calibrating or operating a magnetic sensor for sensor temperature or operating conditions are provided. The magnetic sensor can comprise a dual magnetometer sensor that comprises a first, low-power-consumption magnetometer (e.g., a magneto-inductive magnetometer) and a second higher-power-consumption magnetometer (e.g., a magneto-resistive magnetometer). The second magnetometer can have a lower unit-to-unit variation in temperature calibration parameters and can be used to temperature-correct readings from the first magnetometer. The magnetic sensor can dynamically switch between usage of the first magnetometer and the second magnetometer in order to provide a dynamic sample rate that can depend on conditions within the sensor or external to the sensor.

Abstract: Examples of systems and methods for locating movable objects such as carts (e.g., shopping carts) are disclosed. Such systems and methods can use dead reckoning techniques to estimate the current position of the movable object. Various techniques for improving accuracy of position estimates are disclosed, including compensation for various error sources involving the use of magnetometer and accelerometer, and using vibration analysis to derive wheel rotation rates. Various techniques utilize characteristics of the operating environment in conjunction with or in lieu of dead reckoning techniques, including characteristic of environment such as ground texture, availability of signals from radio frequency (RF) transmitters including precision fix sources.

Abstract: A system for monitoring and controlling shopping cart usage comprises a wheel assembly that attaches to a shopping cart. In some embodiments the wheel assembly includes a wheel, a brake that can be activated to inhibit rotation of the wheel, a controller that controls the brake, a VLF receiver, and an RF transceiver. The RF transceiver may, for example, operate in a 2.4 GHz frequency band. In some implementations, the RF transceiver may be used to detect entry of the shopping cart into a checkout area of the store, and the VLF receiver may be used to detect that the shopping cart is exiting the store. The controller may activate the brake if the shopping cart attempts to exit the store without first passing through a checkout area.

Abstract: A motorized cart retriever, which may be a cart pusher or a cart puller, can apply a force to a nest of human-propelled, wheeled carts to facilitate retrieval of the carts. The cart retriever can include a transceiver configured to wirelessly receive cart status information from cart transmitters of the wheeled carts and wirelessly report event data to a control unit. The cart status information may include an identification of the cart transmitter, a location of the cart, a lock or unlock status of a cart wheel, a misuse condition, etc. The event data can include the cart status information, a number of wheeled carts being retrieved, etc. The cart wheel may include a brake. The transceiver may communicate a message to the cart wheel to keep the brake unactuated during retrieval. The control unit may analyze the event data to detect traffic patterns of the carts.

Abstract: A wheel for a non-motorized vehicle (e.g., a shopping cart) can include a housing assembly and a tread assembly. The housing assembly can be configured to sealingly house electronics or other components. The tread assembly can removably mate with the housing assembly such that the electronics or other components remain closed and/or sealed within the housing assembly when the tread assembly is mated or unmated with the housing assembly.

Abstract: A system for monitoring and controlling shopping cart usage comprises a wheel assembly that attaches to a shopping cart. The wheel assembly includes a wheel, a brake that can be activated to inhibit rotation of the wheel, a controller that controls the brake, a VLF receiver, and an RF transceiver. The RF transceiver may, for example, operate in a 2.4 GHz frequency band. In some implementations, the RF transceiver may be used to detect entry of the shopping cart into a checkout area of the store, and the VLF receiver may be used to detect that the shopping cart is exiting the store. The controller may activate the brake if the shopping cart attempts to exit the store without first passing through a checkout area.

Abstract: A system for monitoring shopping carts uses cameras to generate images of the carts moving in a store. In some implementations, cameras may additionally or alternatively be mounted to the shopping carts and configured to image cart contents. The system may use the collected image data, and/or other types of sensor data (such as the store location at which an item was added to the basket), to classify items detected in the shopping carts. For example, a trained machine learning model may classify item in a cart as “non-merchandise,” “high theft risk merchandise,” “electronics merchandise,” etc. When a shopping cart approaches a store exit without any indication of an associated payment transaction, the system may use the associated item classification data, optionally in combination with other data such as cart path data, to determine whether to execute an anti-theft action, such as locking a cart wheel or activating a store alarm.

Abstract: A system for monitoring shopping carts uses cameras to generate images of the carts moving in a store. In some implementations, cameras may additionally or alternatively be mounted to the shopping carts and configured to image cart contents. The system may use the collected image data, and/or other types of sensor data (such as the store location at which an item was added to the basket), to classify items detected in the shopping carts. For example, a trained machine learning model may classify item in a cart as “non-merchandise,” “high theft risk merchandise,” “electronics merchandise,” etc. When a shopping cart approaches a store exit without any indication of an associated payment transaction, the system may use the associated item classification data, optionally in combination with other data such as cart path data, to determine whether to execute an anti-theft action, such as locking a cart wheel or activating a store alarm.

Abstract: Examples of systems and methods for locating movable objects such as carts (e.g., shopping carts) are disclosed. Such systems and methods can use dead reckoning techniques to estimate the current position of the movable object. Various techniques for improving accuracy of position estimates are disclosed, including compensation for various error sources involving the use of magnetometer and accelerometer, and using vibration analysis to derive wheel rotation rates. Also disclosed are various techniques to utilize characteristics of the operating environment in conjunction with or in lieu of dead reckoning techniques, including characteristic of environment such as ground texture, availability of signals from radio frequency (RF) transmitters including precision fix sources. Such systems and methods can be applied in both indoor and outdoor settings and in retail or warehouse settings.

Abstract: Examples of systems and methods for locating movable objects such as carts (e.g., shopping carts) are disclosed. Such systems and methods can use dead reckoning techniques to estimate the current position of the movable object. Various techniques for improving accuracy of position estimates are disclosed, including compensation for various error sources involving the use of magnetometer and accelerometer, and using vibration analysis to derive wheel rotation rates. Various techniques utilize characteristics of the operating environment in conjunction with or in lieu of dead reckoning techniques, including characteristic of environment such as ground texture, availability of signals from radio frequency (RF) transmitters including precision fix sources.

Abstract: A motorized cart retriever, which may be a cart pusher or a cart puller, can apply a force to a nest of human-propelled, wheeled carts to facilitate retrieval of the carts. The cart retriever can include a transceiver configured to wirelessly receive cart status information from cart transmitters of the wheeled carts and wirelessly report event data to a control unit. The cart status information may include an identification of the cart transmitter, a location of the cart, a lock or unlock status of a cart wheel, a misuse condition, etc. The event data can include the cart status information, a number of wheeled carts being retrieved, etc. The cart wheel may include a brake. The transceiver may communicate a message to the cart wheel to keep the brake unactuated during retrieval. The control unit may analyze the event data to detect traffic patterns of the carts.