Method and apparatus for providing accurate localization for an industrial vehicle
First Claim
1. A method of operating an industrial vehicle in a physical environment, wherein:
- the industrial vehicle comprises a mobile computer and a plurality of sensor devices;
the plurality of sensor devices comprise a wheel encoder, an IMU, or both, and one or more two-dimensional laser scanners;
the wheel encoder, IMU, or both, provide odometry data of the industrial vehicle;
the two-dimensional laser scanner provides details of the physical environment;
the mobile computer comprises an EBN module that employs a priority queue that receives input messages from the plurality of sensor devices and associates each input message with a data source and an acquisition time stamp;
the plurality of sensor devices have different sampling periods and different sampling delays so that an order in which sensor data from the plurality of sensor devices is acquired is not the same as an order in which the sensor data becomes available to the EBN module;
the industrial vehicle is moved along a vehicle path by utilizing an Extended Kalman Filter of the mobile computer to model the position of the industrial vehicle in a two-dimensional plane as a probability density, use the odometry data to update a predicted position of the industrial vehicle, and correct for error in the predicted position of the industrial vehicle using environmental features extracted from the two-dimensional laser scanner by comparing the extracted environmental features with a known map of the physical environment;
the predicted vehicle position update by the Extended Kalman Filter is delayed until a trigger message initiating the vehicle position update is received by the EBN module; and
the EBN module processes the input messages in the priority queue in the order of acquisition time upon availability of the trigger message.
3 Assignments
0 Petitions
Accused Products
Abstract
A method and apparatus for providing accurate localization for an industrial vehicle is described; including processing at least one sensor input message from a plurality of sensor devices, wherein the at least one sensor input message includes information regarding observed environmental features; determining position measurements associated with the industrial vehicle in response to at least one sensor input message, wherein the plurality of sensor devices comprises a two-dimensional laser scanner, and at least one other sensor device selected from an odometer, an ultrasonic sensor, a compass, an accelerometer, a gyroscope, an inertial measurement unit, or an imaging sensor; and updating a vehicle state using the position measurements.
-
Citations
17 Claims
-
1. A method of operating an industrial vehicle in a physical environment, wherein:
-
the industrial vehicle comprises a mobile computer and a plurality of sensor devices; the plurality of sensor devices comprise a wheel encoder, an IMU, or both, and one or more two-dimensional laser scanners; the wheel encoder, IMU, or both, provide odometry data of the industrial vehicle; the two-dimensional laser scanner provides details of the physical environment; the mobile computer comprises an EBN module that employs a priority queue that receives input messages from the plurality of sensor devices and associates each input message with a data source and an acquisition time stamp; the plurality of sensor devices have different sampling periods and different sampling delays so that an order in which sensor data from the plurality of sensor devices is acquired is not the same as an order in which the sensor data becomes available to the EBN module; the industrial vehicle is moved along a vehicle path by utilizing an Extended Kalman Filter of the mobile computer to model the position of the industrial vehicle in a two-dimensional plane as a probability density, use the odometry data to update a predicted position of the industrial vehicle, and correct for error in the predicted position of the industrial vehicle using environmental features extracted from the two-dimensional laser scanner by comparing the extracted environmental features with a known map of the physical environment; the predicted vehicle position update by the Extended Kalman Filter is delayed until a trigger message initiating the vehicle position update is received by the EBN module; and the EBN module processes the input messages in the priority queue in the order of acquisition time upon availability of the trigger message. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15)
-
-
16. A method of operating an industrial vehicle in a physical environment, wherein:
-
the industrial vehicle comprises a mobile computer and a plurality of sensor devices; the plurality of sensor devices comprise a wheel encoder, an IMU, or both, and a plurality of two-dimensional laser scanners; the wheel encoder, IMU, or both, provide odometry data of the industrial vehicle; the two-dimensional laser scanners provide details of the physical environment and are mounted at different measurable positions on the industrial vehicle; the mobile computer comprises an EBN module that transposes laser scan data from the two-dimensional laser scanners to a common reference frame and combines the transposed laser scan data into a single, virtual laser scan; and the EBN module utilizes the single, virtual laser scan to control movement of the industrial vehicle along a vehicle path.
-
-
17. A method of operating an industrial vehicle in a physical environment, wherein:
-
the industrial vehicle comprises a mobile computer and a plurality of sensor devices; the plurality of sensor devices comprise a wheel encoder, an IMU, or both, for providing odometry data of the industrial vehicle; at least one additional sensor device provides details of the physical environment; the mobile computer comprises an EBN module that employs a priority queue that receives input messages from the plurality of sensor devices and associates each input message with a data source and an acquisition time stamp; the plurality of sensor devices have different sampling periods and different sampling delays so that an order in which sensor data from the plurality of sensor devices is acquired is not the same as an order in which the sensor data becomes available to the EBN module; the industrial vehicle is moved along a vehicle path by utilizing an Extended Kalman Filter of the mobile computer to model the position of the industrial vehicle in a two-dimensional plane as a probability density, use the odometry data to update a predicted position of the industrial vehicle, and correct for error in the predicted position of the industrial vehicle using environmental features extracted from the additional sensor device by comparing the extracted environmental features with a known map of the physical environment; the predicted vehicle position update by the Extended Kalman Filter is delayed until a trigger message initiating the vehicle position update is received by the EBN module; and the EBN module processes the input messages in the priority queue in the order of acquisition time upon availability of the trigger message.
-
Specification