Positioning method and device, equipment and storage medium

Abstract

The invention discloses a positioning method and device, equipment and a storage medium, and the method comprises the steps: obtaining a first IMU data set of an IMU device worn by a positioning object, inputting the first IMU data set into a neural network model to obtain a first confidence coefficient, and enabling the first confidence coefficient to represent the consistency of the movement directions of the positioning object and the IMU device, the neural network model is obtained through training based on a second IMU data set of the IMU device and a second confidence coefficient obtained in the movement process of the positioning object wearing the IMU device, and the second confidence coefficient is a true value representing the consistency of the movement direction of the positioning object and the movement direction of the IMU device; and determining the position of the positioning object according to the first confidence. The invention aims to improve the positioning accuracy.

Claims

1. A positioning method, It is characterized in that The method comprises: Acquire a first IMU data set of an IMU device worn by a positioning object, wherein there is a relative motion between the IMU device and the positioning object during the process of the positioning object wearing the IMU device; Inputting the first IMU data set into a neural network model to obtain a first confidence level, wherein the first confidence level represents the consistency of the movement direction of the positioning object and the IMU device, and the neural network model is trained based on a second IMU data set of the IMU device acquired during the movement of the positioning object wearing the IMU device and a second confidence level, wherein the second confidence level is a true value representing the consistency of the movement direction of the positioning object and the IMU device; The position of the positioning object is determined according to the first confidence level.

2. The method according to claim 1, It is characterized in that The method further comprises: Acquire the second IMU data set and the position data set of the positioning object during the movement of the positioning object wearing the IMU device; The second confidence level is determined based on the IMU data in the second IMU data set and the position data in the position data set at the same time.

3. The method according to claim 2, It is characterized in that Acquiring the second IMU data set and the position data set of the positioning object includes: Acquire the second IMU data set of the IMU device in N time intervals, and the position data set of the positioning object in the N time intervals, where there is at least one time unit in each time interval, the interval between adjacent time units is the time interval for collecting data, and N is an integer greater than or equal to 1; Determining the second confidence level according to the IMU data in the second IMU data set and the position data in the position data set at the same time includes: Aligning the IMU data and the position data corresponding to the same time unit in the same time interval, the IMU data belonging to the second IMU data set, and the position data belonging to the position data set; Determine N second confidences according to the aligned second IMU data set and the position data set in the N time intervals; The method further comprises: The neural network model is trained according to the second IMU data set in the N time intervals and the N second confidence levels.

4. The method according to claim 3, It is characterized in that The aligning of the IMU data and the position data corresponding to the same time unit in the same time interval includes: The IMU data and the position data corresponding to the same time unit in each time interval are aligned according to the timestamps included in the second IMU data set and the position data set.

5. The method according to claim 3, It is characterized in that The determining N second confidences according to the aligned second IMU data set and the position data set in the N time intervals includes: Determine N first direction sets and N second direction sets according to the aligned IMU data and the position data in the N time intervals, wherein the first direction sets and the second direction sets correspond to each other one by one, the first direction set includes the movement direction of the IMU device in all time units in any time interval obtained based on the IMU data, and the second direction set includes the movement direction of the positioning object in all time units in any time interval obtained based on the position data; The N second confidences are determined according to the N first direction sets and the N second direction sets.

6. The method according to claim 5, It is characterized in that Determining the N second confidences according to the N first direction sets and the N second direction sets includes: Calculate the difference between a motion direction corresponding to any time unit in the first direction set and a motion direction corresponding to any time unit in the second direction set; Calculate, according to the difference, an average value of the difference between all movement directions in the first direction set and all corresponding movement directions in the second direction set in any time interval; Calculate the variance based on the difference and the average value; The variance is normalized to obtain the second confidence level.

7. The method according to claim 5, It is characterized in that Determining the N second confidences according to the N first direction sets and the N second direction sets includes: Calculate the difference between a motion direction corresponding to any time unit in the first direction set and a motion direction corresponding to any time unit in the second direction set; Calculate the gradient of the difference corresponding to adjacent time units in any time interval according to the difference; The gradient is normalized to obtain the second confidence.

8. The method according to any one of claims 1 to 7, It is characterized in that Determining the position of the positioning object according to the first confidence level includes: When the first confidence level is greater than or equal to a preset threshold, the position of the IMU device is determined to be the position of the positioning object.

9. A positioning device, It is characterized in that The device comprises: An acquisition module is used to acquire a first IMU data set of an IMU device worn by a positioning object, wherein there is a relative motion between the IMU device and the positioning object during the process of the positioning object wearing the IMU device; A calculation module, configured to input the first IMU data set into a neural network model to obtain a first confidence level, wherein the first confidence level represents the consistency of the movement direction of the positioning object and the IMU device, and the neural network model is trained based on a second IMU data set of the IMU device acquired during the movement of the positioning object wearing the IMU device and a second confidence level, wherein the second confidence level is a true value representing the consistency of the movement direction of the positioning object and the IMU device; A determination module is used to determine the position of the positioning object according to the first confidence level.

10. A positioning device, It is characterized in that The positioning device comprises: a memory, a processor and a program stored in the memory for implementing the positioning method. The memory is used to store a program for implementing the positioning method; The processor is used to execute a program for implementing the positioning method, so as to implement the steps of the positioning method according to any one of claims 1 to 8.

11. A storage medium, It is characterized in that The storage medium stores a program for implementing the positioning method, and the program for implementing the positioning method is executed by a processor to implement the steps of the positioning method as claimed in any one of claims 1 to 8.

Images