Unmanned ship track control accuracy calibration method

Abstract

The invention relates to a method for calibrating track control accuracy of an unmanned ship, which comprises the following steps of: firstly, mounting a master station of a wireless Beidou positioning system on a river bank, and mounting a slave station of the Beidou positioning system and a three-axis acceleration sensor on a deck of the unmanned ship; real-time measurement of the space position and the navigation speed of the unmanned ship is realized between the master station and the slave station through a differential positioning technology; then, the track and the propulsion power of the unmanned ship are set, and the unmanned ship is made to sail autonomously according to a specific sailing route; according to the method, the measurement error of the positioning system is reduced by using the iterative smooth filtering technology, the difference between the unmanned ship track and the preset track is scientifically represented by using the matrix norm for the first time, and finally the unmanned ship track control accuracy is measured by using the correlation coefficient of the two tracks.

Description

Technical Field:

[0001] The present invention belongs to the technical field of metrology and testing, and particularly relates to a calibration method for the accuracy of the track control of an unmanned ship. Background Art:

[0002] An unmanned ship is an unmanned platform used for ocean observation, transportation, and maritime military struggle. It has functions such as positioning, navigation, autonomous decision-making, and control, and works in a remote control, pre-programmed, or autonomous manner. At the same time, by carrying a variety of different ocean observation sensors, the unmanned ship can automatically correct and plan the navigation route according to the measurement results of the sensors during the voyage. Different from the automatic driving of a car, the autonomous navigation of a ship is more susceptible to external environmental interference, and factors such as water flow, wind, and waves can significantly affect the ship's heading. In recent years, the rapid development of artificial intelligence and sensing technology has significantly improved the intelligent level of unmanned ships. In the inland rivers and harbors with complex waterways, advanced unmanned ships can all perform autonomous navigation.

[0003] The precise control of the track of an unmanned ship is the basis for realizing the autonomous navigation of an unmanned vehicle. Before the advent of global positioning technology, the automatic control of ships was limited to the control of the heading and could not achieve path tracking and planning. After the satellite navigation systems with differential positioning technology such as GPS and Beidou became mature, the path control methods of ships developed rapidly. Researchers have successively proposed track control methods such as optimal control, variable structure control, feedback linearization, intelligent control, Backstepping algorithm, and model predictive control, which have solved technical problems such as the nonlinearity, uncertainty, stability, and time delay of the ship navigation system, and can adaptively adjust the propulsion rate and rudder direction of the ship under environmental interference conditions, ensuring the accuracy of the autonomous navigation track of the unmanned ship to the greatest extent.

[0004] However, affected by water flow and wind and waves, it is very difficult for an unmanned vehicle to completely ensure that it travels on the preset track. The degree of track deviation, that is, the accuracy of track control, is a key parameter for measuring the intelligent level of an unmanned ship. At present, the spatial Euclidean distance between the target position and the advancing position is usually used to measure the accuracy of track control. Theoretically, when the number of target positions is large, the connecting line of multiple advancing positions can also approximately represent the track of the unmanned ship. At this time, the measurement result of the track control error is more accurate. When the track of the unmanned ship deviates, there is a cumulative error in track control, and the control error sizes of tracks of different lengths are different, so it is very difficult to establish a unified evaluation standard. In addition, the measurement of the track completely depends on the real-time measurement data of the positioning system. Factors such as the random error and system error of the positioning system cause the track measurement to be inconsistent with the actual track. Especially for the spatial positioning based on the differential principle, the spatial position calculation requires the spatial position averaging in seconds, further reducing the accuracy of track measurement. Summary of the Invention:

[0005] The technical problem to be solved by the present invention is to provide a calibration method for the accuracy of the track control of an unmanned ship. This method uses iterative smoothing filtering technology to reduce the measurement error of the positioning system, and for the first time proposes to scientifically characterize the difference between the track of the unmanned ship and the preset track using matrix norms, and finally measures the accuracy of the track control of the unmanned ship using the correlation coefficient between the two tracks.

[0006] The technical solution of the present invention is to provide a calibration method for the accuracy of the track control of an unmanned ship: First, install the master station of the wireless Beidou positioning system on the river bank, and install the slave station of the Beidou positioning system and a 3-axis acceleration sensor on the deck of the unmanned ship. The real-time measurement of the spatial position and navigation speed of the unmanned ship is realized between the master and slave stations through differential positioning technology; then, set the track and propulsion power of the unmanned ship, and let the unmanned ship navigate autonomously along a specific navigation route; use the measurement results of the Beidou positioning system and the acceleration sensor to form the state variable X k ; construct a measurement model, establish an iterative equation, update the measurement results of the track of the unmanned ship, and process the measurement results of the spatial position through iterative smoothing filtering technology; respectively construct the spatial position matrices S = [s1, s2,..., s N and W = [w1, w2,..., w N of the preset track and the actual track of the unmanned ship, and calculate the norms of the matrices S, W, and SW T respectively; finally, obtain the accuracy of the track control of the unmanned ship as

[0007] Preferably, the Beidou positioning system can measure the geodetic coordinates (x, y, h) of the unmanned ship, and the 3-axis acceleration sensor can measure the speeds v x , v y , v z of the three axes of the unmanned ship.

[0008] Preferably, the state variable X k is composed of the geodetic coordinates measured by the Beidou positioning system and the speeds of the three axes. X k is a vector containing 6 elements, and the specific expression is as follows:

[0009] X k = (x k , y k , h k , v xk , v yk , v zk ) T .

[0010] Preferably, an iterative model is established, and the spatial information of the unmanned ship at the (k + 1)-th positioning can be expressed as

[0011] Xk+1 = AX k + q k

[0012] wherein,

[0013]

[0014] q k obeys a normal distribution with a mean of 0 and a variance of Q

[0015]

[0016] The measurement model is

[0017] Y k = HX k + q k

[0018] wherein,

[0019]

[0020] q k obeys a normal distribution with a mean of 0 and a variance of .

[0021] Preferably, according to the measurement results, the steps for measuring the spatial position of the unmanned ship based on the iterative smoothing filtering principle are as follows:

[0022] (1) Calculate the mean value of a single measurement and the variance

[0023]

[0024] (2) Calculate the gain of the iterative step size as

[0025]

[0026] (3) According to the measurement results of the spatial point positions, calculate the updated measurement mean m k and the variance P k as

[0027]

[0028] (4) Calculate the spatial position of the unmanned ship after iterative smoothing filtering as

[0029]

[0030] Preferably, the preset spatial trajectory S of the unmanned ship is

[0031]

[0032] Then ||S|| = Tr(SS T ), where Tr represents the trace of a matrix.

[0033] The actual measured trajectory W is

[0034]

[0035] Then ||W|| = Tr(WW T ).

[0036] Therefore, the accuracy of the unmanned ship trajectory control can be calculated by the following formula

[0037]

[0038] Compared with the prior art, the present invention has the following advantages:

[0039] The present invention proposes a calibration method for the accuracy of the unmanned ship trajectory control based on iterative smoothing filtering, which reduces the measurement error of the positioning system by using iterative smoothing filtering technology, first proposes to use matrix norm to scientifically characterize the difference between the unmanned ship trajectory and the preset trajectory, and finally measures the accuracy of the unmanned ship trajectory control by using the correlation coefficient of the two trajectories. Description of the Drawings:

[0040] Figure 1 It is a schematic diagram of the device for calibrating the accuracy of the unmanned ship trajectory of the present invention.

[0041] Figure 2 It is a comparison diagram of the iterative smoothing filtering effect of the unmanned ship trajectory of the present invention. Detailed Embodiments:

[0042] The following further describes the present invention in detail with reference to the drawings:

[0043] The method for calibrating the accuracy of the unmanned ship trajectory of the present invention includes the following steps:

[0044] (1) Install the Beidou positioning system and the 3-axis acceleration sensor in the manner shown Figure 1 , where the master station of the Beidou positioning system is installed on the shore and can provide a stable reference system for the differential positioning system; the slave station of the Beidou positioning system is installed at the tail of the unmanned ship, and it should be ensured that the slave station is not blocked by metal objects to prevent affecting the reception of radio signals; the 3-axis acceleration sensor is fixedly connected to the stern to facilitate accurate measurement of the speed of the unmanned ship. Among them, the Beidou positioning system can measure the geodetic coordinates (x, y, h) of the unmanned ship, and the 3-axis acceleration sensor can measure the speeds v x , v y , v z of the unmanned ship in three axial directions.

[0045] The method for the acceleration sensor to measure the speeds of the unmanned ship in the x, y, and z directions is as follows

[0046]

[0047] (2) Through the unmanned ship control software, set the propulsion power and the track of the unmanned ship. Use the Beidou positioning system to time the 3-axis acceleration sensor so that the measured data of the two are synchronized in time. Through the communication system of the unmanned ship, feedback the measurement results of the 3-axis acceleration sensor to the computer in real time.

[0048] (3) Construct the state variables of the spatial position of the unmanned ship. The state variable X of the spatial position of the unmanned ship k consists of the geodetic coordinates measured by the Beidou positioning system and the speeds of the 3 axes. X k is a vector containing 6 elements, and the specific expression is as follows

[0049] X k =(x k , y k , h k , v xk , v yk , v zk ) T

[0050] (4) Establish an iterative model. The spatial information of the unmanned ship at the (k + 1)-th positioning can be expressed as

[0051] X k+1 = AX k + q k

[0052] where

[0053]

[0054] q k obeys the normal distribution with a mean of 0 and a variance of Q

[0055]

[0056] The measurement model is

[0057] Y k = HX k + q k

[0058] where

[0059]

[0060] q k obeys the normal distribution with a mean of 0 and a variance of .

[0061] (4) Perform iterative smoothing filtering, and the specific steps are as follows:

[0062] i) Calculate the mean value of single measurement and variance

[0063]

[0064] ii) Calculate the gain of the iterative step size as

[0065]

[0066] iii) According to the measurement results of the spatial points, calculate the updated measurement mean value m k and variance P k as

[0067]

[0068] iv) Calculate the spatial position of the unmanned ship after iterative smoothing filtering as

[0069]

[0070] (5) According to the preset track of the unmanned ship and the track after iterative smoothing filtering, calculate the track accuracy of the unmanned ship. The preset spatial track S of the unmanned ship is

[0071]

[0072] Then ||S|| = Tr(SS T ), where Tr represents the trace of the matrix.

[0073] The actual measured track W is

[0074]

[0075] Then ||W|| = Tr(WW T ).

[0076] Therefore, the track control accuracy of the unmanned ship can be calculated by the following formula.

[0077]

[0078] (6) The initial position is (0, 1, 0), the initial velocity is (0, 1, 0), the time sampling interval Δt = 0.1 s, q x = q y = q z = 1, When, the preset track, the measurement by the Beidou positioning system and the track after iterative smoothing filtering are as Figure 2As shown in the figure. It can be seen that the iterative smoothing filtering technology effectively eliminates the measurement errors of the Beidou positioning system. At this time, the accuracy ε of the track control of the unmanned ship is 0.912.

[0079] The present invention proposes a calibration method for the accuracy of the track control of an unmanned ship based on iterative smoothing filtering. On the basis of the measurements of the Beidou positioning system and the acceleration sensor, the spatial position and attitude information are fused, and the iterative smoothing filtering technology is used to smooth the spatial position measurement data, reducing the measurement errors of the positioning system.

[0080] The above is only an illustration of the preferred embodiments of the present invention, but it should not be construed as a limitation of the claims. Any equivalent process transformation using the specification of the present invention is included in the scope of patent protection of the present invention.

Claims

1. A calibration method for the accuracy of the track control of an unmanned ship, characterized in that: Including the following steps, First, install the master station of the wireless Beidou positioning system on the river bank, and install the slave station of the Beidou positioning system and a 3-axis acceleration sensor on the deck of the unmanned ship. The real-time measurement of the spatial position and navigation speed of the unmanned ship is realized between the master and slave stations through differential positioning technology; Then, set the track and propulsion power of the unmanned ship, and let the unmanned ship navigate autonomously along a specific navigation route; Next, the measurement results of the Beidou positioning system and the acceleration sensor are used to form the state variable X k ; construct a measurement model, establish an iterative equation, update the measurement results of the unmanned ship's track, and process the measurement results of the spatial position through iterative smoothing filtering technology; And, respectively construct the spatial position matrices of the preset track and the actual track of the unmanned ship S = [s1, s2, …, s N , W = [w1, w2, …, w N , respectively calculate the norms of the matrices S, W, and SW T Finally, the track control accuracy of the unmanned ship is obtained as 2. The method for calibrating the accuracy of the track control of an unmanned ship according to claim 1, characterized in that: The Beidou positioning system can measure the geodetic coordinates (x, y, h) of the unmanned ship, and the three-axis acceleration sensor can measure the velocities v x 、v y 、v z 。 3. The method for calibrating the accuracy of the unmanned ship's track control according to claim 1, characterized in that: State variable X of the spatial position of the unmanned ship k It consists of the geodetic coordinates measured by the Beidou positioning system and the velocities of 3 axes, X k It is a vector containing 6 elements, and the specific expression is as follows:

4. The method for calibrating the accuracy of the unmanned ship's track control according to claim 2, wherein: The spatial information of the unmanned ship at the (k + 1)-th positioning can be expressed as X k+1 = AX k + q k where, q k obeys a normal distribution with a mean of 0 and a variance of Q The measurement model is Y k = HX k + q k where, q k obeys a normal distribution with a mean of 0 and a variance of .

5. The method for calibrating the accuracy of the track control of an unmanned ship according to claim 1, wherein: The measurement steps for the spatial position through the iterative smoothing filtering technology are (1) Calculate the mean of a single measurement and the variance (2) Calculate the gain of the iterative step size as (3) Calculate and update the measurement mean m based on the measurement results of the spatial points k and the variance P k as (4) Calculate the spatial position of the unmanned ship after iterative smoothing filtering as 6. The method for calibrating the accuracy of the track control of an unmanned ship according to claim 1, characterized in that: The preset spatial trajectory S is Then ||S|| = Tr(SS T ), where Tr represents the trace of a matrix; The actual measured trajectory W is Then ||W|| = Tr(WW T ); The track control accuracy of the unmanned ship can be calculated by the following formula,

Images