Self-adaptive time-distance ship track prediction method based on mirror gate network

Abstract

To provide a self-adaptive time-distance ship track prediction method based on a mirror gate network.SOLUTION: A method comprises the following steps: building the mirror gate network; extracting equal-time-distance track feature points; constructing input track data; and constructing a variable-length time interval adaptive track prediction model. A novel mirror image gate network is established, mirror image gates are trained from the two ends of data, and training of the track prediction model is completed by adjusting the time interval when the error obtained by the model meets a convergence condition.EFFECT: The two-end learning capability of the mirror network enables the model to spontaneously discover and utilize hidden mode in the data to give full play to the autonomous learning and adaptive capability of the model, so as to obtain higher track prediction precision.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to the technical field of ship wake prediction, and in particular to a self-adaptive time interval ship wake prediction method based on mirror gate network. [Background technology]

[0002] In recent years, with the continuous development of artificial intelligence technology, deep learning has made great strides in the field of ship trajectory prediction. Representative trajectory prediction methods include artificial neural networks (ANNs), recurrent neural networks (RNNs), long short-term memory networks (LSTMs), and gated recurrent units (GRUs). Furthermore, temporal convolutional networks (TCNs) have attracted increasing attention due to their advantages in time series modeling and feature capture. Subsequently, the proposed attention mechanism became a new trend, and this cutting-edge method has gradually been applied to the field of trajectory prediction, leading to numerous variants. The introduction of attention mechanisms allows the model to help extract important information from the data.

[0003] Patent Documents 1 and 2 disclose a prediction model that uses a loss function to learn the model in order to minimize errors in trajectory prediction. However, it is difficult to capture the dependencies and dynamic patterns before and after the trajectory, and a loss function is required to evaluate the merits and demerits of the prediction model, resulting in problems such as generalized learning and a lack of adaptability. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Chinese Patent Application Publication No. 117093889 [Patent Document 2] Chinese Patent Application Publication No. 116629116 Summary of the Invention [Problem to be solved by the invention]

[0005] The present invention has been made in consideration of the above-mentioned problems existing in the prior art, and aims to propose a ship wake prediction method with a self-adaptive time interval based on a mirror gate network that can understand and capture dependencies and dynamic patterns in time series data. The present invention is the first to propose neural network technology based on mirror learning, which employs a method of self-learning the model and self-checking the results, moving in both directions until the mirrors from both ends overlap, thereby completing the training of the prediction model without evaluating the loss function. [Means for solving the problem]

[0006] In order to achieve the above object, the solution of the present invention is as follows.

[0007] A method for predicting ship trajectories with self-adaptive time intervals based on a mirror gate network, which is one embodiment of the present invention, includes the following steps S1 to S4: S1, the process of constructing a mirror gate network; S2, a step of extracting trajectory feature points at equal time intervals; S3, constructing input track data; and S4, constructing a self-adaptive track prediction model with variable long-time interval; In S1, the mirror gate network comprises a forward gate unit and a backward gate unit; S1 includes the following steps S11 to S14: S11, setting the forward gate unit of the mirror gate network; In S11, the forward gate unit includes a forward merging gate, a forward diverging gate, and a forward regenerating gate, and the forward gate unit is used to control the input and output of track information; S11 includes the following steps S111 to S113, S111, the forward merging gate is the current track information input state x t and the unit status of the previous gate unit

number

number

number

number

number

[0008] S112, the forward branch gate converts the information passed through the forward merge gate into the unit status of the previous gate unit.

number

number

number

number

number

number

number

[0009] S113, the forward regeneration gate converts the information passed through the forward merging gate into the current track information input state x t and the unit state of the current unit

number

number

number

number

number

number

number

[0010] S12, setting the reverse direction gate unit of the mirror gate network; In S12, the backward gate unit includes a backward merging gate, a backward diverting gate, and a backward regenerating gate, and the backward gate unit performs a backward operation on the input information; S12 includes the following steps S121 to S123: S121, the reverse direction merging gate is the current track information input state x t and the unit status of the next gate unit

number

number

number

number

number

[0011] S122, the reverse diversion gate converts the information passed through the reverse merging gate into the unit state of the next gate unit according to the following formula:

number

number

number

number

number

number

number

[0012] S123, the reverse regeneration gate converts the information passed through the reverse merging gate into the current track information input state x t and the unit state of the current unit

number

number

number

number

number

number

number

[0013] S13, the forward gate units are sequentially connected in the forward direction, and the backward gate units are sequentially connected in the backward direction, and the states of the forward gate units and the backward gate units at the same time interval are merged to form the mirror gate network, and the calculation formula is as follows:

number

number

[0014] S14, perform calculations using the mirror gate network to obtain the following predicted output information:

number

number

[0015] S2 includes the following steps S21 to S24: S21, by extracting feature points of self-adaptive time intervals, the track P in the AIS data collected in the same sea area k is defined and expressed as a track feature vector, and the calculation formula is as follows: P k ={p1 k ,p2 k ,...,p i k ,...,p l k} where p i k is the track P of the kth ship in the sea area at timestamp i. k represents the track point of p i k The characteristics of are expressed by the following formula: p i k =(lon ik ,lat i k ,cog i k ,sog i k ) where lon i k represents the longitude value of the ship, and lat i k represents the latitude value of the ship, and cog i k represents the ship's course over the ground, and sog i k represents the ship's ground speed, The self-adaptation time interval is defined as t and expressed as follows: t=(t1,t2,...,t i ,...,t n ) where t i is P k represents the time distance of the i-th track point in the

[0016] S22, interval [t i ,t i+1 ], we define an interpolation function expressed by the following formula, X i (t i )=a i t i 3 +b i t i 2 +c i t i +d i X i+1 (t i+1 )=a i t i+1 3 +b i t i+1 2 +c i t i+1 +d i Taking the first derivative of the above function, we get the following equation: X i (t i )′=3a i t i 2+2b i t i +c i X i+1 (t i+1 )′=3a i t i+1 2 +2b i t i+1 +c i Combining the interpolation function and its derivative and solving the equation gives a i , b i , c i and d i Find the value of .

[0017] S23, X i (t i )' and X i+1 (t i+1 )′, where all intermediate points of the interpolating function are calculated as M j (t j ,x j ) and M according to the following formula: j The first-order difference quotient of the left and right coordinate points of k j and k j+1 It is expressed as

number

number

number

[0018] S24: The time series obtained by equally dividing the track recording period is substituted into an interpolation function of longitude and latitude in each period to obtain feature points extracted at equal time intervals of the track.

[0019] S3 includes the following steps S31 to S32: S31: In order to retain the time series characteristics, dynamic patterns and correlation information of the track data and allow the model to better learn the change / evolution process of the track data, a moving time window process is performed to establish a moving time window, where the length of the moving time window is set as L and the moving distance is set as L / q, where q is a flexibly selected integer, and the smaller q is, the stronger the correlation between the data is.

[0020] S32, the AIS data that has undergone feature extraction and moving time window processing at self-adaptive time intervals is input to the mirror gate network, and the forward gate unit and the backward gate unit are respectively trained, and one or more trajectories are trained and predicted by the mirror gate network, and the input data I at time t is used. t and output data O t is expressed by the following formula:

number

[0021] In S4, each input data is trained through N forward gate units and N backward gate units, and the self-adaptive variable length N is determined based on the final result of the mirror gate network, and the optimal trajectory prediction result is obtained. [Effects of the Invention]

[0022] The present invention has the following beneficial effects compared to the prior art.

[0023] 1. This invention constructs a new mirror gate network, which trains the mirror gates from both ends of the data and adjusts the time distance. When the error obtained by the model meets the convergence condition, the trajectory prediction model is trained. This new gate structure deeply captures the dependencies and dynamic patterns before and after the trajectory.

[0024] 2. The mirror network constructed by the present invention has the ability to learn from both ends, allowing the model to spontaneously discover and utilize hidden modes in the data, fully utilizing its autonomous learning and adaptive capabilities, thereby achieving higher trajectory prediction accuracy.

[0025] 3. The present invention uses the method of self-adaptive time interval feature extraction to achieve self-adaptive alignment of different track data, and can also construct input data for multiple track prediction. [Brief explanation of the drawings]

[0026] [Figure 1] FIG. 1 is a schematic flow chart of the present invention. [Figure 2] 1 is a schematic diagram showing a forward gate and a reverse gate; [Figure 3] FIG. 1 is a schematic diagram illustrating a mirror gate network. [Figure 4] FIG. 10 shows a schematic diagram of the extraction of equitemporally spaced AIS points. [Figure 5] FIG. 10 is a diagram illustrating the construction of an input track. [Figure 6] FIG. 1 is a diagram illustrating a trajectory prediction model. DETAILED DESCRIPTION OF THE INVENTION

[0027] We will further describe the present invention in combination with the following drawings: As shown in Figure 1, the self-adaptive time interval ship wake prediction method based on Miller gate network includes the following steps:

[0028] S1: Construct a mirror gate network. Figure 2 shows the specific structure of the forward gate and the reverse gate. Figure 3 shows the details of the mirror gate network constructed based on the forward gate and the reverse gate. This structure is used as a basic model.

[0029] S2. Aligning the data at equal time intervals. As shown in Figure 4, data alignment is achieved by constructing interpolation functions in four different dimensions: lon, lat, cog, and sog.

[0030] S3: Construct input data. As shown in Figure 5, the aligned track data undergoes a moving time window process with a length of L and a stride length of L / q, and various track data are merged to construct the input for the multiple track prediction model.

[0031] S4: Construct a self-adaptive track prediction model with a variable long-term interval. As shown in Figure 6, train N forward and backward gate networks, stop training when the loss of the forward and backward gates is minimized, and determine the self-adaptive variable length N according to the final result.

[0032] The present invention is not limited to the present embodiment, and any equivalent ideas or modifications within the technical scope disclosed in the present invention are included in the scope of protection of the present invention.

[0033] (Addendum) (Appendix 1) A step S1 of constructing a mirror gate network; A step S2 of extracting trajectory characteristic points at equal time intervals; a step S3 of constructing input track data; and and a step S4 of constructing a self-adaptive track prediction model with a variable long time interval; In S1, the mirror gate network comprises a forward gate unit and a backward gate unit; S1 includes the following steps S11 to S14: S11, setting the forward gate unit of the mirror gate network; In S11, the forward gate unit includes a forward merging gate, a forward diverging gate, and a forward regenerating gate, and the forward gate unit is used to control the input and output of track information; S11 includes the following steps S111 to S113, S111, the forward merging gate is the current track information input state x t and the unit status of the previous gate unit

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

number

Claims

[Claim 1] Step S1 of constructing a mirror gate network; A step S2 of extracting track characteristic points at equal time intervals; a step S3 of constructing input track data; and and (S4) constructing a self-adaptive track prediction model with a variable long time interval; In S1, the mirror gate network comprises a forward gate unit and a backward gate unit; S1 includes the following steps S11 to S14: S11, setting the forward gate unit of the mirror gate network; In S11, the forward gate unit includes a forward merging gate, a forward diverting gate, and a forward regenerating gate, and the forward gate unit is used to control the input and output of track information; S11 includes the following steps S111 to S113: S111, the forward merging gate is in the current track information input state x ｔ and the unit status of the previous gate unit [Equation 1] and the hidden layer state of the previous gate unit [Equation 2] The forward merge gate is responsible for aggregating all input information, and the calculation formula is as follows: [Equation 3] where: [Equation 4] are respectively [Equation 5] represents the weight of , sigmoid represents the activation function, S112, the forward branch gate transmits the information passed through the forward merge gate to the unit status of the previous gate unit. [Equation 6] and the hidden layer state of the previous gate unit [Equation 7] A branch is formed along with the new unit state [Equation 8] and the hidden layer state [Equation 9] The calculation formula is as follows: [Equation 10] where: [0011] are respectively [0012] represents the weight corresponding to the process of calculating tanh, and tanh represents the activation function. S113, the forward regeneration gate converts the information passed through the forward merging gate into the current track information input state x ｔ and the unit state of the current unit [0013] and hidden layer state [0014] With new condition [Equation 15] The calculation formula is as follows: [0016] where: [Equation 17] are respectively [Equation 18] In the process of generating [Equation 19] represents the weight corresponding to S12, setting the reverse gate unit of the mirror gate network; In S12, the backward gate unit includes a backward merging gate, a backward dividing gate, and a backward regenerating gate, and the backward gate unit performs a backward operation on the input information; S12 includes the following steps S121 to S123: S121, the reverse direction merging gate is the current track information input state x ｔ and the unit status of the next gate unit [Equation 20] and the hidden layer state of the next gate unit [0000] The backward merging gate is responsible for aggregating all input information, and the calculation formula is as follows: [Equation 22] where: [Equation 23] are respectively [0000] represents the weight of S122, the reverse diversion gate converts the information passed through the reverse merging gate into the unit state of the next gate unit according to the following formula: [Equation 25] and the hidden layer state of the next gate unit [Equation 26] A branch is formed along with the new unit state [0000] and the hidden layer state [0000] The calculation formula is as follows: [0000] where: [Equation 30] are calculated in the process of calculating the reverse shunt gates. [Equation 31] represents the weight corresponding to S123, the reverse regeneration gate converts the information passed through the reverse merging gate into the current track information input state x ｔ and the unit state of the current unit [Equation 32] and hidden layer state [Equation 33] With new condition [Equation 34] The calculation formula is as follows: [Equation 35] where: [Equation 36] are respectively [Equation 37] In the process of generating [Equation 38] represents the weight corresponding to S13, the forward gate units are sequentially connected in the forward direction, and the backward gate units are sequentially connected in the backward direction, and the states of the forward gate units and the backward gate units at the same time interval are merged to form the mirror gate network, and the calculation formula is as follows: [Number 39] where: [Equation 40] represents the merge operation, S14: Perform calculations using the mirror gate network to obtain the following predicted output information: [Equation 41] Here, w ｇｙ teeth, [0.001] represents the weight of S2 includes the following steps S21 to S24: S21, by extracting the feature points of the self-adaptive time interval, the track P in the AIS data collected in the same sea area ｋ is defined and expressed as a track feature vector, and the calculation formula is as follows: P ｋ ={p １ ｋ ,p ２ ｋ ,...,p ｉ ｋ ,...,p ｌ ｋ } Here, p ｉ ｋ is the track P of the kth ship in the sea area at timestamp i. ｋ represents the track point of ｉ ｋ The characteristics of are expressed by the following formula: p ｉ ｋ =(lon ｉ ｋ , lat ｉ ｋ , cog ｉ ｋ ,sog ｉ ｋ ) Here, lon ｉ ｋ represents the longitude value of the ship, and lat ｉ ｋ represents the latitude value of the ship, and cog ｉ ｋ represents the ship's course over the ground, and sog ｉ ｋ represents the ship's ground speed, The self-adaptation time interval is defined as t and expressed as follows: t=(t １ ,t ２ ,...,t ｉ ,...,t ｎ ) Here, t ｉ Is, P ｋ represents the time distance of the i-th track point in S22, section [t ｉ , t ｉ＋１ ], an interpolation function expressed by the following formula is defined: X ｉ (t ｉ )=a ｉ t ｉ ３ +b ｉ t ｉ ２ +c ｉ t ｉ +d ｉ X ｉ＋１ (t ｉ＋１ )=a ｉ t ｉ＋１ ３ +b ｉ t ｉ＋１ ２ +c ｉ t ｉ＋１ +d ｉ Taking the first derivative of the above function, we get the following equation: X ｉ (t ｉ )′=3a ｉ t ｉ ２ +2b ｉ t ｉ +c ｉ X ｉ＋１ (t ｉ＋１ )′=3a ｉ t ｉ＋１ ２ +2b ｉ t ｉ＋１ +c ｉ Combining the interpolation function and its derivative and solving the equation gives a ｉ , b ｉ , c ｉ and d ｉ Find the value of S23, X ｉ (t ｉ )' and X ｉ＋１ (t ｉ＋１ )', where all intermediate points of the interpolating function are calculated using M ｊ (t ｊ , x ｊ ) and M according to the following formula: ｊ The first-order difference quotient of the left and right coordinate points of k ｊ Tok ｊ＋１ It is expressed as [Equation 43] According to the following formula, the midpoint M ｊ The weights of the two adjacent periods on the left and right of are ω ｊ and ω ｊ＋１ and [Equation 44] According to the following formula, the midpoint M ｊ The first derivative in X ｊ (t)' is approximated, [Equation 45] S24: Substituting the time series obtained by equally dividing the track recording period into an interpolation function of longitude and latitude in each period, to obtain feature points extracted at equal time intervals of the track; S3 includes the following steps S31 to S32: S31: Perform a moving time window process to establish a moving time window to retain the time series characteristics, dynamic patterns, and correlation information of the track data and allow the model to better learn the change / evolution process of the track data, and set the length of the moving time window as L and the moving distance as L / q, where q is a flexibly selected integer, and the smaller q is, the stronger the correlation between the data before and after; S32, the AIS data that has undergone feature extraction and moving time window processing for self-adaptive time intervals is input to the mirror gate network, and the forward gate unit and the backward gate unit are trained respectively, and one or more trajectories are trained and predicted by the mirror gate network, and the input data I at time t is ｔ and output data O ｔ is expressed by the following formula: [Equation 46] where r represents the number of predicted tracks and S ｔ ｋ represents the track data of the kth ship processed at time t, and lon ｔ ｋ represents the interpolated longitude at time t, and lat ｔ ｋ represents the interpolated latitude at time t, and cog ｔ ｋ represents the course corresponding to the interpolated latitude and longitude at time t, and sog ｔ ｋ represents the navigation speed corresponding to the interpolated latitude and longitude at time t, L represents the length of the moving time window at time t, and u represents the length of the predicted trajectory of the model. In S4, each input data is trained through N forward gate units and N backward gate units, and a self-adaptive variable length N is determined according to the final result of the mirror gate network, and the optimal trajectory prediction result is obtained; A self-adaptive time interval ship wake prediction method based on mirror gate network, characterized by:

Images