CN115086375A - Method, device, system and medium for compensating motion state information delay of networked vehicle - Google Patents
Method, device, system and medium for compensating motion state information delay of networked vehicle Download PDFInfo
- Publication number
- CN115086375A CN115086375A CN202210668381.5A CN202210668381A CN115086375A CN 115086375 A CN115086375 A CN 115086375A CN 202210668381 A CN202210668381 A CN 202210668381A CN 115086375 A CN115086375 A CN 115086375A
- Authority
- CN
- China
- Prior art keywords
- delay
- information
- motion state
- state information
- moment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application provides a method, a device, a system and a medium for compensating motion state information delay of networked vehicles, wherein the method comprises the following steps: the method comprises the steps of obtaining current motion state information and current delay information of a sending vehicle, obtaining historical motion state information and corresponding historical delay information within preset time of the sending vehicle, inputting the current motion state information, the current delay information, the historical motion state information and the historical delay information into a long-short term memory network model, enabling the long-short term memory network model to output prediction compensation information of the sending vehicle, and calculating according to the prediction compensation information to obtain prediction compensation displacement of the sending vehicle. Therefore, according to the long-short term memory network model, the vehicle motion state information and the delay information can be modeled and predicted to compensate the vehicle motion state information in the time period of vehicle-mounted communication delay, so that the aim of reducing the influence of vehicle-mounted communication delay is fulfilled, and the running safety of the networked vehicle is improved.
Description
Technical Field
The present application relates to the field of computers, and in particular, to a method, an apparatus, a system, and a medium for compensating for a motion state information delay of a networked vehicle.
Background
In the past decades, the appearance of the technology of internet-Connected Autonomous Vehicles (CAV) brings a new revolution to the transportation system, and is beneficial to significantly improving the experience of our daily driving in the aspects of safety, mobility, sustainability and the like.
The networked automatic driving vehicles equipped with communication equipment can run cooperatively, the cooperation among the vehicles mainly means that the vehicle-to-vehicle (V2V), the vehicle-to-road (V2I), the vehicle-to-person (V2P), the vehicle-to-network (V2N) and the like can realize the omnibearing communication between the vehicle and the surrounding vehicles, environment and network through the vehicle networking V2X (vehicle-to-observing) technology, and the environment-to-vehicle-to-network (V2N) provides environment sensing, information interaction and cooperative control capability for the vehicle driving and traffic management application.
Vehicles rely on-board sensing sensors, such as cameras, radar, and lidar, to measure the status of neighboring vehicles. With the introduction of V2X communication, the networked autonomous vehicle can obtain data beyond its direct sensing range and obtain information that cannot be detected by remote sensors, which helps to improve the sensing range of the networked autonomous vehicle. However, in terms of vehicle communication technology, problems such as communication delay are inevitably introduced, which will degrade the performance of any networked autonomous vehicle application.
That is, after the receiving vehicle receives the motion state information sent by the sending vehicle, the sending vehicle has a certain displacement within the delay time, and this vehicle communication delay problem causes that the information (position, speed, acceleration, etc.) of the sending vehicle acquired by the receiving vehicle is not real-time, and the safety of the receiving vehicle is certainly affected.
Disclosure of Invention
In view of the above, an object of the present application is to provide a method, an apparatus, a system and a medium for compensating for delay of motion state information of a networked vehicle, which can reduce the influence of vehicle-mounted communication delay and improve the safety of operation of the networked vehicle.
In order to achieve the purpose, the technical scheme is as follows:
in a first aspect, an embodiment of the present application provides a method for compensating for delay of online vehicle motion state information, including:
acquiring current motion state information and current time delay information of a sending vehicle; the current motion state information comprises a current position, a current speed and a current acceleration; the current delay information comprises a delay time from the previous moment corresponding to the current moment;
acquiring historical motion state information and corresponding historical delay information within preset time of the sending vehicle; the historical motion state information comprises positions at all moments, speeds at all moments and accelerations at all moments; the historical delay information comprises delay time lengths respectively corresponding to each moment and the previous moment of each moment;
inputting the current motion state information, the current delay information, the historical motion state information and the historical delay information into a long-short term memory network model so that the long-short term memory network model outputs the prediction compensation information of the sending vehicle; the prediction compensation information comprises the speed after prediction delay, the acceleration after prediction delay and the prediction delay time;
and calculating the predicted compensation displacement of the sending vehicle according to the predicted compensation information.
In one possible implementation, before the inputting the current motion state information, the current delay information, the historical motion state information, and the historical delay information into a long-short term memory network model, the method further includes:
obtaining a training set of the long-short term memory network model, the training set comprising: actual motion state information at a known moment, actual delay information at the known moment, actual motion state information at a next moment at the known moment, and actual delay information at the next moment at the known moment;
the actual motion state information at the known moment comprises an actual position, an actual speed and an actual acceleration at the known moment; the actual delay information of the known moment comprises the actual delay time from the last moment corresponding to the known moment; the actual motion state information of the next moment of the known moment comprises an actual position, an actual speed and an actual acceleration of the next moment of the known moment; the actual delay information of the next moment of the known moment comprises the actual delay time from the known moment to the next moment of the known moment;
learning the mapping relation between the motion state information of the known moment and the actual delay information of the known moment and the actual motion state information of the next moment of the known moment and the actual delay information of the next moment of the known moment by utilizing the training set;
and determining model parameters of the long-term and short-term memory network model according to the mapping relation.
In one possible implementation, inputting the current motion state information, the current delay information, the historical motion state information, and the historical delay information into a long-short term memory network model includes:
encoding the current motion state information, the current delay information, the historical motion state information and the historical delay information and then inputting the encoded current motion state information, the current delay information, the historical motion state information and the historical delay information into the long-short term memory network model;
and the coding comprises the step of carrying out unified standardized processing on the current motion state information, the current delay information, the historical motion state information and the historical delay information.
In one possible implementation manner, the calculating the predicted compensation displacement of the sending vehicle according to the predicted compensation information includes:
the predicted compensated displacement of the sending vehicle is equal to the product of the predicted delayed speed and the predicted delay period.
In a second aspect, an embodiment of the present application provides a device for compensating for delay of motion state information of an internet-connected vehicle, including:
the first acquisition unit is used for acquiring the current motion state information and the current time delay information of the sending vehicle; the current motion state information comprises a current position, a current speed and a current acceleration; the current delay information comprises a delay time from the previous moment corresponding to the current moment;
the second acquisition unit is used for acquiring historical motion state information and corresponding historical delay information within preset time of the sending vehicle; the historical motion state information comprises positions at all moments, speeds at all moments and accelerations at all moments; the historical delay information comprises delay time lengths from each moment to the last moment of each moment, which respectively correspond to the delay time lengths;
an input unit, configured to input the current motion state information, the current delay information, the historical motion state information, and the historical delay information into a long-short term memory network model, so that the long-short term memory network model outputs prediction compensation information of the sending vehicle; the prediction compensation information comprises the speed after prediction delay, the acceleration after prediction delay and the prediction delay time;
and the calculation unit is used for calculating the predicted compensation displacement of the sending vehicle according to the predicted compensation information.
In one possible implementation, the apparatus further includes:
a third obtaining unit, configured to obtain a training set of the long-term and short-term memory network model, where the training set includes: actual motion state information at a known moment, actual delay information at the known moment, actual motion state information at a next moment at the known moment, and actual delay information at the next moment at the known moment;
the actual motion state information at the known moment comprises an actual position, an actual speed and an actual acceleration at the known moment; the actual delay information of the known moment comprises the actual delay time from the last moment corresponding to the known moment; the actual motion state information of the next moment of the known moment comprises an actual position, an actual speed and an actual acceleration of the next moment of the known moment; the actual delay information of the next moment of the known moment comprises the actual delay time from the known moment to the next moment of the known moment;
a learning unit, configured to learn, by using the training set, a mapping relationship between the motion state information at the known time and the actual delay information at the known time, and the actual motion state information at the next time of the known time and the actual delay information at the next time of the known time;
and the determining unit is used for determining the model parameters of the long-term and short-term memory network model according to the mapping relation.
In a possible implementation manner, the input unit is specifically configured to:
encoding the current motion state information, the current delay information, the historical motion state information and the historical delay information and then inputting the encoded current motion state information, the current delay information, the historical motion state information and the historical delay information into the long-short term memory network model;
and the coding comprises the step of carrying out unified standardized processing on the current motion state information, the current delay information, the historical motion state information and the historical delay information.
In a possible implementation manner, the computing unit is specifically configured to:
the predicted compensated displacement of the sending vehicle is equal to the product of the predicted delayed speed and the predicted delay period.
In a third aspect, an embodiment of the present application provides a system for compensating for delay of online vehicle motion state information, including:
a memory for storing a computer program;
and the processor is used for realizing the steps of the online vehicle motion state information delay compensation method when executing the computer program.
In a fourth aspect, the present application provides a computer-readable medium, where a computer program is stored on the computer-readable medium, and when the computer program is processed and executed, the steps of the method for compensating for delay of online vehicle motion state information are implemented as described above.
Compared with the prior art, the embodiment of the application has the following beneficial effects:
the embodiment of the application provides a method, a device, a system and a medium for compensating motion state information delay of a networked vehicle, wherein the method comprises the following steps: acquiring current motion state information and current delay information of a sending vehicle, wherein the current motion state information comprises a current position, a current speed and a current acceleration, and the current delay information comprises a delay time from a previous moment to a current moment corresponding to the current moment; obtaining historical motion state information and corresponding historical delay information within preset time of a sending vehicle, wherein the historical motion state information comprises positions at all moments, speeds at all moments and accelerations at all moments, and the historical delay information comprises delay time from each moment to the last moment at each moment, which respectively corresponds to the delay time; and inputting the current motion state information, the current delay information, the historical motion state information and the historical delay information into the long-short term memory network model so that the long-short term memory network model outputs the predicted compensation information of the sending vehicle, wherein the predicted compensation information comprises the predicted delayed speed, the predicted delayed acceleration and the predicted delay time, and the predicted compensation displacement of the sending vehicle is obtained through calculation according to the predicted compensation information. Therefore, according to the long-short term memory network model, the vehicle motion state information and the delay information can be modeled and predicted to compensate the vehicle motion state information in the time period of vehicle-mounted communication delay, so that the aim of reducing the influence of vehicle-mounted communication delay is fulfilled, and the running safety of the networked vehicle is improved.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
The above and other features, advantages and aspects of various embodiments of the present disclosure will become more apparent by referring to the following detailed description when taken in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numbers refer to the same or similar elements. It should be understood that the drawings are schematic and that elements and features are not necessarily drawn to scale.
Fig. 1 shows a flowchart of a method for compensating for delay of online vehicle motion state information according to an embodiment of the present application;
fig. 2 is a schematic diagram illustrating devices included in an application scenario according to an embodiment of the present application;
FIG. 3 is a schematic diagram illustrating node units of an LSTM model according to an embodiment of the present disclosure;
fig. 4 shows a flowchart of a method for compensating for delay of online vehicle motion state information according to an embodiment of the present application;
FIG. 5 is a schematic diagram illustrating a relationship between historical information and predicted future information under a network architecture according to an embodiment of the present application;
FIG. 6 is a diagram illustrating the number of neurons in each layer of the LSTM model provided by an embodiment of the present application;
fig. 7 shows a schematic diagram of a networked vehicle motion state information delay compensation device provided by an embodiment of the present application.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present application more comprehensible, embodiments accompanying the present application are described in detail below with reference to the accompanying drawings.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application, but the present application may be practiced in other ways than those described herein, and it will be apparent to those of ordinary skill in the art that the present application is not limited by the specific embodiments disclosed below.
As described in the background art, the advent of the technology of networked Autonomous vehicles (CAV) has brought a new revolution to transportation systems in the past decades, contributing to a significant improvement in our daily driving experience in terms of safety, mobility, sustainability, and the like.
The networked automatic driving vehicles equipped with communication equipment can run cooperatively, and the cooperation among the vehicles mainly means that the vehicle-to-vehicle, environment and network all-around communication can be realized through the vehicle networking V2X (vehicle-to-vehicle) technology, and the vehicle-to-vehicle, V2V, vehicle-to-infrastructure, V2I, vehicle-to-pedestrian (V2P), vehicle-to-network, V2N and the like, so that the environment sensing, information interaction and cooperative control capability is provided for the vehicle driving and traffic management application.
Vehicles rely on-board sensing sensors, such as cameras, radar, and lidar, to measure the status of neighboring vehicles. With the introduction of V2X communication, the networked autonomous vehicle can obtain data beyond its direct sensing range and obtain information that cannot be detected by the remote sensor, which helps to improve the sensing range of the networked autonomous vehicle. However, in terms of vehicle communication technology, problems such as communication delay are inevitably introduced, which will degrade the performance of any networked autonomous vehicle application.
That is, after the receiving vehicle receives the motion state information sent by the sending vehicle, the sending vehicle has a certain displacement within the delay time, and this vehicle communication delay problem causes that the information (position, speed, acceleration, etc.) of the sending vehicle acquired by the receiving vehicle is not real-time, and the safety of the receiving vehicle is certainly affected.
The applicant researches and finds that researchers in different directions do researches in different angles on how to reduce the influence of vehicle-mounted communication delay:
the communication direction focuses on reducing the influence of time delay from the viewpoints of communication protocol optimization or route optimization and the like: from the point of protocol optimization, a service channel time slot medium access control mechanism can be designed to improve the channel utilization rate, increase the throughput and reduce the time delay of the channel; from the perspective of route optimization, the routing algorithm can be improved, and the link failure time can be predicted by using the information such as node position, movement speed and the like, so as to improve the performances in the aspects of end-to-end delay of data packets, transmission throughput rate, message delivery rate and the like. Although handling delays has been good from a communication perspective, the delay and packet loss problems in V2X communications are not permanently eliminated from the underlying communication mechanism.
The control direction focuses on the application of designing the controller based on a physical motion model, the controller is provided with a feedback unit and a motion estimation unit, the motion state of the vehicle can be compensated according to real-time errors, and meanwhile, the influence of the communication delay on the overall stability and safety can be researched by considering the condition of the communication delay in the design process of the controller. However, such a motion model is not reliable for long-term prediction, and the vehicle trajectory tends to be non-linear due to the driver's decision during driving, and such a model tends to be less than perfect for dealing with the non-linear situation.
For example, some examples of the application of delay compensation for on-board communication with respect to control direction are motion estimation of a vehicle using a dynamic model method, such as an extended kalman filter, an unscented kalman filter, and a particle filter. These filters are modified versions of kalman filters, which are based on a first order markov chain, i.e. the state of the current time step depends only on the state of the previous time step. Although they are useful for real-time implementation, a key limitation is that the sensor data prior to the previous time step cannot be used directly. And the performance of, for example, a kalman filter depends on the accuracy of the parameter matrices, in particular the process noise covariance matrix Q and the measurement noise covariance matrix R. In practice, the choice of Q and R plays an important role in the evaluation of the kalman filter, since the measurement noise is device-dependent, and thus different hardware platforms have different noise characteristics, and the theoretical derivation of these covariance matrices may not be accurate enough for all platforms.
The deep learning direction focuses on modeling analysis and prediction of data transmitted by vehicle-mounted communication directly, can avoid the problem of noise modeling and is concentrated on the data. Meanwhile, the problem of nonlinearity which cannot be solved by controlling the direction can be effectively solved. BP (Back Propagation) model and RNN (Recurrent Neural Network) model are commonly used. The BP neural network model processes input data by dividing an input layer, a hidden layer, and an output layer. The weights in the neural network are continuously updated and optimized through a feedback intermediary. The output data approaches the actual value with any precision so as to achieve the purpose of prediction. However, the problem of the BP neural network in the field of vehicle trajectory prediction is that the number of input features is fixed, and historical trajectory data cannot be used. While the BP neural network is based on an assumption that the input data is uniformly distributed and independent. However, in order to calculate the motion state information of the vehicle, time-series data that needs to be input are correlated. Therefore, the vehicle motion state prediction model can be designed by using the recurrent neural network RNN, and the memory information from the previous moment is added on the basis of the input of the current moment, so that the model has certain memory, but the RNN has the defect that the gradient disappears.
In order to solve the above technical problem, an embodiment of the present application provides a method, an apparatus, a system, and a medium for compensating for a delay of a vehicle motion state information of an internet, where the method includes: acquiring current motion state information and current delay information of a sending vehicle, wherein the current motion state information comprises a current position, a current speed and a current acceleration, and the current delay information comprises a delay time from a previous moment to a current moment corresponding to the current moment; obtaining historical motion state information and corresponding historical delay information within preset time of a sending vehicle, wherein the historical motion state information comprises positions at all moments, speeds at all moments and accelerations at all moments, and the historical delay information comprises delay time from each moment to the last moment at each moment, which respectively corresponds to the delay time; and inputting the current motion state information, the current delay information, the historical motion state information and the historical delay information into the long-short term memory network model so that the long-short term memory network model outputs the predicted compensation information of the sending vehicle, wherein the predicted compensation information comprises the predicted delayed speed, the predicted delayed acceleration and the predicted delay time, and the predicted compensation displacement of the sending vehicle is obtained through calculation according to the predicted compensation information. Therefore, according to the long-short term memory network model, the vehicle motion state information and the delay information can be modeled and predicted to compensate the vehicle motion state information in the time period of vehicle-mounted communication delay, so that the aim of reducing the influence of vehicle-mounted communication delay is fulfilled, and the running safety of the networked vehicle is improved.
Exemplary method
Referring to fig. 1, a flowchart of a method for compensating for delay of information of a motion state of a networked vehicle according to an embodiment of the present application includes:
s101: acquiring current motion state information and current time delay information of a sending vehicle; the current motion state information comprises a current position, a current speed and a current acceleration; the current delay information comprises a delay time length from the last moment corresponding to the current moment.
In the embodiment of the application, the current motion state information and the current time delay information of the sending vehicle can be obtained; the current motion state information may include a current position, a current velocity, and a current acceleration; the current delay information may include a delay time from a previous time to a current time corresponding to the current time.
Specifically, the data may come from a Basic Safety Messages (BSM) and delay information sent by the sending vehicle through the v2x communication device, where the BSM information includes key motion state information such as the current position, current speed, and current acceleration of the sending vehicle at the sending time. The vehicle motion state information, such as road scenes, vehicle speed and the like, changes within a certain range, so that the prediction of the future motion state information of the vehicle by using the historical motion state information of the vehicle becomes possible.
It should be noted that the position information in the BSM is latitude and longitude height information, which does not change much within a certain time, and the latitude and longitude information cannot be directly fit into the model for calculation, and coordinate transformation is required, so that redundancy of the algorithm is increased, and therefore the method does not directly use historical position information for trajectory prediction, but focuses on predicting motion state variables such as speed and acceleration for delay compensation.
For example, a possible application scenario provided in the embodiment of the present application may be, as shown in fig. 2, the scenario includes a data processing terminal 101 of a sending vehicle j; the v2x communication device 102 of the sending vehicle j; a v2x communication device 103 of the receiving vehicle i and a data processing terminal 104 of the receiving vehicle i.
The sending vehicle j sends the motion state information at the time t, the sending vehicle j has been displaced at the time t to t +1, and the information about the sending vehicle j in the receiving vehicle i stays at the time before t and t, so that delay compensation is needed to compensate the information about the sending vehicle j in the receiving vehicle i at the time t to t + 1.
The v2x communication device 103 of the receiving vehicle i receives the BSM information about the sending vehicle j at the time k from the v2x communication device 102 of the sending vehicle j, the BSM information contains key information such as position, speed, acceleration and the like, and records a delay value from the last time to the current time, and then carries out delay compensation on the motion information about the vehicle j in the data processing terminal 104 of the receiving vehicle i.
Optionally, the current motion state information and the current delay information may be subjected to information preprocessing, and then, data required for neural network training is subjected to packing processing, and specifically, the packed current motion state information and current delay information set may be:
X j (t)＝{v j (t)，a j (t)，d j (t)}；
wherein v is j (t) is the current speed, a j (t) is the current acceleration, d j And (t) is the time delay from the previous moment corresponding to the current moment.
S102: acquiring historical motion state information and corresponding historical delay information within preset time of the sending vehicle; the historical motion state information comprises positions at all moments, speeds at all moments and accelerations at all moments; the historical delay information comprises delay time lengths from each moment to the last moment of each moment, which respectively correspond to the delay time lengths.
In the embodiment of the application, historical motion state information and corresponding historical delay information within a preset time length of a sending vehicle can be obtained; the historical motion state information comprises the position, the speed and the acceleration of each moment; the historical delay information includes delay durations corresponding to the previous time from each time.
Specifically, X j (t-1) may be a set of historical motion state information and corresponding historical delay information at time t-1, X j (t-1)＝{v j (t-1)，a j (t-1)，d j (t-1) }; wherein v is j (t-1) is the speed at the time of history t-1, a j (t-1) acceleration at time t-1 of history, d j And (t-1) is the time delay from the historical time t-1 to the time t-2.
Thus, a total set comprising current motion state information, current delay information, historical motion state information, and corresponding historical delay information may be formed:
X＝{X j (t)、X j (t-1)...}。
it should be noted that the speed and acceleration information herein have two directions, namely, a lateral direction and a longitudinal direction, for calculating the lateral displacement and the longitudinal displacement of the vehicle, and for convenience of description, the two directions are represented by a single symbol.
S103: inputting the current motion state information, the current delay information, the historical motion state information and the historical delay information into a long-short term memory network model so that the long-short term memory network model outputs the prediction compensation information of the sending vehicle; the prediction compensation information comprises the speed after prediction delay, the acceleration after prediction delay and the prediction delay time length.
S104: and calculating the predicted compensation displacement of the sending vehicle according to the predicted compensation information.
In the embodiment of the present application, the current motion state information, the current delay information, the historical motion state information, and the historical delay information may be input into a Long Short-Term Memory network model (LSTM).
In particular, reference is made to FIG. 3, which is provided for an embodiment of the present applicationA structure diagram of an LSTM cell node, at each time step t, g (x) t ) Is the input of the LSTM subunit, h t Is the output of the LSTM unit node. Outputting h to the unit node of the last time step t-1 Input g (x) with the current system t ) In combination, the input for the current cell is formed. The state of each unit node is C t It records the system memory. C t There is an update in each time step. To control the flow of information through the unit nodes, several gates are applied, including an input gate (i) t ) Output gate (o) t ) And forget gate (f) t ). Each gate generates an output between 0 and 1, where the value of the output is calculated by a sigmoid (σ) function. An output of 0 indicates that the input to the gate is completely blocked, while an output of 1 indicates that all the information of the input is held in the cell node. The input, output and forget gate calculation method is as follows:
i t ＝σ(W i g(x t )+U i h t-1 +b i )
o t ＝σ(W o g(x t )+U o h t-1 +b o )
f t ＝σ(W f g(x t )+U f h t-1 +b f )；
wherein, W i ，W o And W f The weights for three gates; u shape i ，U o And U f Is the corresponding cyclic weight; b i ，b o And b f Is the offset value of three gates.
Similar to the AND gate function, combine the current input with the previous cell state C t-1 The cell state is updated. The difference is that the input will be processed by a hyperbolic tangent function, instead of sigmoid, which generates an output between-1 and 1:
after the update, the user can use the data to update the data,
multiplied by the output of the input gate and then used as the first component to update the state of the cell. Another component for updating the cell state is the previous cell state, which is processed by the forget gate to determine how to use the past data. For both components, the cell state at t will be updated as:
unit h to be used at t +1 t Is calculated by multiplication of the output gate with the tanh function of the current cell state:
h t ＝o t tanh(c t )。
in a possible implementation manner, in order to improve the accuracy of model prediction, a semi-supervised learning strategy may be adopted to obtain a training set of the long-term and short-term memory network model, where the training set includes: actual motion state information at a known time, actual delay information at a known time, actual motion state information at a next time at a known time, and actual delay information at a next time at a known time.
The actual motion state information at the known moment comprises the actual position, the actual speed and the actual acceleration at the known moment; the actual delay information of the known moment comprises the actual delay time from the last moment corresponding to the known moment; the actual motion state information at the next moment of the known moment comprises an actual position, an actual speed and an actual acceleration at the next moment of the known moment; the actual delay information of the next moment of the known moment comprises the actual delay time from the known moment to the next moment of the known moment;
learning the mapping relation between the motion state information at the known moment and the actual delay information at the known moment and the actual motion state information at the next moment of the known moment and the actual delay information at the next moment of the known moment by using a training set;
and determining model parameters of the long-term and short-term memory network model according to the mapping relation, so that the state information and the delay information at a plurality of historical moments are used as training data, and the state information at the next moment is used as a data label for training, thereby improving the accuracy of model prediction.
In one possible implementation, referring to fig. 4, in order to improve the prediction stability and performance of the long-short term memory network, the current motion state information, the current delay information, the historical motion state information, and the historical delay information may be encoded and input into the long-short term memory network model.
Specifically, the encoding includes performing unified standardization processing on the current motion state information, the current delay information, the historical motion state information, and the historical delay information. And then, training and using a network model, which is helpful for improving the stability and performance of the network, and encoding can be performed by adopting an encoder.
In addition, the decoder can be used for carrying out data de-standardization according to the related coding of the coder so as to achieve the purpose of data scaling.
Optionally, the following criteria are used:
wherein
Is the nth component of the input data, e.g. delay, velocity or acceleration at time t, i.e. x t，n Normalized input of, mu N 、σ N Is the mean and variance from the total sample.
Referring to fig. 5, a graph of the relationship between historical information and predicted future information under the network architecture of the method provided by the embodiment of the present application is shown, where each time T is represented t-h+1 To T t The Input at each moment is Input into the LSTM model after being coded, and finally decoded and output to predict the future T t+1 Prediction compensation information Output at time.
The length of the input sequence, i.e., the historical time step, is an important factor that affects the predictive performance. According to the preliminary experiments of the inventors, the length is preferably 5 to 15.
It should be noted that, the sequences used in the neural network training of the present method are of equal step size, but the motion state information in the sequences is the result of changes after several unequal delay values occur, which is also one of the reasons why the trajectory prediction is not directly performed by using the position information.
Specifically, for the LSTM model, as shown in fig. 6, Regression prediction (Regression) is performed from the Input Layer (Input Layer) to the Full connection Layer (Full connection Layer) to the LSTM Layer to the Full connection Layer. The number of neurons per sublayer is shown in parentheses.
Optionally, taking the uniform variable motion as an example, a detailed description of the compensation motion state information is made.
The predicted delayed state information is: x j (t+1)＝{v j (t+1)，a j (t+1)，d j (t+1)}
Wherein, X j (t +1) is a delayed information set, v j (t +1) is the predicted delayed speed, a j (t +1) is the predicted acceleration, d j (t +1) is the predicted delay value from time t to time t + 1.
Specifically, the predicted compensation displacement Δ p of the sending vehicle is equal to the product of the predicted delayed speed and the predicted delay time.
Δp＝v j (t+1)*d j (t+1)；
Predicting the speed after the delay: v. of j (t+1)＝a j (t+1)*d j (t+1)。
The embodiment of the application provides a method for compensating motion state information delay of a networked vehicle, which comprises the following steps: acquiring current motion state information and current delay information of a sending vehicle, wherein the current motion state information comprises a current position, a current speed and a current acceleration, and the current delay information comprises a delay time from a previous moment to a current moment corresponding to the current moment; obtaining historical motion state information and corresponding historical delay information within a preset time length of a sending vehicle, wherein the historical motion state information comprises positions at all moments, speeds at all moments and accelerations at all moments, and the historical delay information comprises delay time lengths from all the moments to the last moment at all the moments, which correspond to each other; and inputting the current motion state information, the current delay information, the historical motion state information and the historical delay information into the long-short term memory network model so that the long-short term memory network model outputs the predicted compensation information of the sending vehicle, wherein the predicted compensation information comprises the predicted delayed speed, the predicted delayed acceleration and the predicted delay time, and the predicted compensation displacement of the sending vehicle is obtained through calculation according to the predicted compensation information. Therefore, according to the long-short term memory network model, the vehicle motion state information and the delay information can be modeled and predicted to compensate the vehicle motion state information in the time period of vehicle-mounted communication delay, so that the aim of reducing the influence of vehicle-mounted communication delay is fulfilled, and the running safety of the networked vehicle is improved.
Exemplary devices
Referring to fig. 2, a device for compensating for motion state information delay of a networked vehicle according to an embodiment of the present application includes:
a first obtaining unit 201, configured to obtain current motion state information and current delay information of a sending vehicle; the current motion state information comprises a current position, a current speed and a current acceleration; the current delay information comprises a delay time from the previous moment corresponding to the current moment;
a second obtaining unit 202, configured to obtain historical motion state information and corresponding historical delay information within a preset time duration of the sending vehicle; the historical motion state information comprises positions at all moments, speeds at all moments and accelerations at all moments; the historical delay information comprises delay time lengths from each moment to the last moment of each moment, which respectively correspond to the delay time lengths;
an input unit 203, configured to input the current motion state information, the current delay information, the historical motion state information, and the historical delay information into a long-short term memory network model, so that the long-short term memory network model outputs prediction compensation information of the sending vehicle; the prediction compensation information comprises the speed after prediction delay, the acceleration after prediction delay and the prediction delay time;
a calculating unit 204, configured to calculate a predicted compensation displacement of the sending vehicle according to the predicted compensation information.
In one possible implementation, the apparatus further includes:
a third obtaining unit, configured to obtain a training set of the long-term and short-term memory network model, where the training set includes: actual motion state information at a known moment, actual delay information at the known moment, actual motion state information at a next moment at the known moment, and actual delay information at the next moment at the known moment;
the actual motion state information at the known moment comprises an actual position, an actual speed and an actual acceleration at the known moment; the actual delay information of the known moment comprises the actual delay time from the last moment corresponding to the known moment; the actual motion state information of the next moment of the known moment comprises an actual position, an actual speed and an actual acceleration of the next moment of the known moment; the actual delay information of the next moment of the known moment comprises the actual delay time from the known moment to the next moment of the known moment;
a learning unit, configured to learn, by using the training set, a mapping relationship between the motion state information at the known time and the actual delay information at the known time, and the actual motion state information at the next time of the known time and the actual delay information at the next time of the known time;
and the determining unit is used for determining the model parameters of the long-term and short-term memory network model according to the mapping relation.
In a possible implementation manner, the input unit is specifically configured to:
encoding the current motion state information, the current delay information, the historical motion state information and the historical delay information and then inputting the encoded current motion state information, the current delay information, the historical motion state information and the historical delay information into the long-short term memory network model;
and the coding comprises the step of carrying out unified standardized processing on the current motion state information, the current delay information, the historical motion state information and the historical delay information.
In a possible implementation manner, the computing unit is specifically configured to:
the predicted compensated displacement of the sending vehicle is equal to the product of the predicted delayed speed and the predicted delay period.
The embodiment of the application provides a device for compensating motion state information delay of a networked vehicle, and a method utilizing the device comprises the following steps: acquiring current motion state information and current delay information of a sending vehicle, wherein the current motion state information comprises a current position, a current speed and a current acceleration, and the current delay information comprises a delay time from a previous moment to a current moment corresponding to the current moment; obtaining historical motion state information and corresponding historical delay information within preset time of a sending vehicle, wherein the historical motion state information comprises positions at all moments, speeds at all moments and accelerations at all moments, and the historical delay information comprises delay time from each moment to the last moment at each moment, which respectively corresponds to the delay time; and inputting the current motion state information, the current delay information, the historical motion state information and the historical delay information into the long-short term memory network model so that the long-short term memory network model outputs the predicted compensation information of the sending vehicle, wherein the predicted compensation information comprises the predicted delayed speed, the predicted delayed acceleration and the predicted delay time, and the predicted compensation displacement of the sending vehicle is obtained through calculation according to the predicted compensation information. Therefore, according to the long-short term memory network model, the vehicle motion state information and the delay information can be modeled and predicted to compensate the vehicle motion state information in the time period of vehicle-mounted communication delay, so that the aim of reducing the influence of vehicle-mounted communication delay is fulfilled, and the running safety of the networked vehicle is improved.
On the basis of the above embodiments, the embodiment of the present application provides a system for compensating for delay of vehicle motion state information of an internet vehicle, including:
a memory for storing a computer program;
and the processor is used for realizing the steps of the online vehicle motion state information delay compensation method when executing the computer program.
On the basis of the foregoing embodiments, the present application further provides a computer-readable medium, where a computer program is stored, and when the computer program is processed and executed, the steps of the above-mentioned networked vehicle motion state information delay compensation method are implemented.
It should be noted that the computer readable medium in the present disclosure can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In contrast, in the present disclosure, a computer readable signal medium may comprise a propagated data signal with computer readable program code embodied therein, either in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, optical cables, RF (radio frequency), etc., or any suitable combination of the foregoing.
The computer readable medium may be embodied in the system described above; or may exist separately and not be assembled into the system.
In particular, according to an embodiment of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a non-transitory computer readable medium, the computer program containing program code for performing the method illustrated by the flow chart.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the apparatus embodiment, since it is substantially similar to the method embodiment, it is relatively simple to describe, and reference may be made to some descriptions of the method embodiment for relevant points.
The foregoing is merely a preferred embodiment of the present application and, although the present application discloses the foregoing preferred embodiments, the present application is not limited thereto. Those skilled in the art can now make numerous possible variations and modifications to the disclosed embodiments, or modify equivalent embodiments, using the methods and techniques disclosed above, without departing from the scope of the claimed embodiments. Therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present application still fall within the protection scope of the technical solution of the present application without departing from the content of the technical solution of the present application.
Claims (10)
1. A method for compensating the motion state information delay of a networked vehicle is characterized by comprising the following steps:
acquiring current motion state information and current time delay information of a sending vehicle; the current motion state information comprises a current position, a current speed and a current acceleration; the current delay information comprises a delay time from the previous moment corresponding to the current moment;
acquiring historical motion state information and corresponding historical delay information within preset time of the sending vehicle; the historical motion state information comprises positions at all moments, speeds at all moments and accelerations at all moments; the historical delay information comprises delay time lengths from each moment to the last moment of each moment, which respectively correspond to the delay time lengths;
inputting the current motion state information, the current delay information, the historical motion state information and the historical delay information into a long-short term memory network model so that the long-short term memory network model outputs the prediction compensation information of the sending vehicle; the prediction compensation information comprises the speed after prediction delay, the acceleration after prediction delay and the prediction delay time;
and calculating the predicted compensation displacement of the sending vehicle according to the predicted compensation information.
2. The method of claim 1, wherein before entering the current kinematic state information, the current latency information, the historical kinematic state information, and the historical latency information into a long-short term memory network model, the method further comprises:
obtaining a training set of the long-short term memory network model, the training set comprising: actual motion state information at a known moment, actual delay information at the known moment, actual motion state information at a next moment at the known moment, and actual delay information at the next moment at the known moment;
the actual motion state information at the known moment comprises an actual position, an actual speed and an actual acceleration at the known moment; the actual delay information of the known moment comprises the actual delay time from the last moment corresponding to the known moment; the actual motion state information of the next moment of the known moment comprises an actual position, an actual speed and an actual acceleration of the next moment of the known moment; the actual delay information of the next moment of the known moment comprises the actual delay time from the known moment to the next moment of the known moment;
learning the mapping relation between the motion state information of the known moment and the actual delay information of the known moment and the actual motion state information of the next moment of the known moment and the actual delay information of the next moment of the known moment by utilizing the training set;
and determining model parameters of the long-term and short-term memory network model according to the mapping relation.
3. The method of claim 1, wherein inputting the current motion state information, the current latency information, the historical motion state information, and the historical latency information into a long-short term memory network model comprises:
encoding the current motion state information, the current delay information, the historical motion state information and the historical delay information and then inputting the encoded current motion state information, the current delay information, the historical motion state information and the historical delay information into the long-short term memory network model;
and the coding comprises the step of carrying out unified standardized processing on the current motion state information, the current delay information, the historical motion state information and the historical delay information.
4. The method of claim 1, wherein said calculating a predicted compensated displacement of the sending vehicle from the predicted compensation information comprises:
the predicted compensated displacement of the sending vehicle is equal to the product of the predicted delayed speed and the predicted delay period.
5. A networked vehicle motion state information delay compensation device is characterized by comprising:
the first acquisition unit is used for acquiring the current motion state information and the current time delay information of the sending vehicle; the current motion state information comprises a current position, a current speed and a current acceleration; the current delay information comprises a delay time from the previous moment corresponding to the current moment;
the second acquisition unit is used for acquiring historical motion state information and corresponding historical delay information within preset time of the sending vehicle; the historical motion state information comprises positions at all moments, speeds at all moments and accelerations at all moments; the historical delay information comprises delay time lengths from each moment to the last moment of each moment, which respectively correspond to the delay time lengths;
an input unit, configured to input the current motion state information, the current delay information, the historical motion state information, and the historical delay information into a long-short term memory network model, so that the long-short term memory network model outputs prediction compensation information of the sending vehicle; the prediction compensation information comprises the speed after prediction delay, the acceleration after prediction delay and the prediction delay time;
and the calculation unit is used for calculating the predicted compensation displacement of the sending vehicle according to the predicted compensation information.
6. The apparatus of claim 5, further comprising:
a third obtaining unit, configured to obtain a training set of the long-term and short-term memory network model, where the training set includes: actual motion state information at a known moment, actual delay information at the known moment, actual motion state information at a next moment at the known moment, and actual delay information at the next moment at the known moment;
the actual motion state information at the known moment comprises an actual position, an actual speed and an actual acceleration at the known moment; the actual delay information of the known moment comprises the actual delay time from the last moment corresponding to the known moment; the actual motion state information of the next moment of the known moment comprises an actual position, an actual speed and an actual acceleration of the next moment of the known moment; the actual delay information of the next moment of the known moment comprises the actual delay time from the known moment to the next moment of the known moment;
a learning unit, configured to learn, by using the training set, a mapping relationship between the motion state information at the known time and the actual delay information at the known time, and the actual motion state information at the next time of the known time and the actual delay information at the next time of the known time;
and the determining unit is used for determining the model parameters of the long-term and short-term memory network model according to the mapping relation.
7. The apparatus of claim 5, wherein the input unit is specifically configured to:
encoding the current motion state information, the current delay information, the historical motion state information and the historical delay information and then inputting the encoded current motion state information, the current delay information, the historical motion state information and the historical delay information into the long-short term memory network model;
and the coding comprises the step of carrying out unified standardized processing on the current motion state information, the current delay information, the historical motion state information and the historical delay information.
8. The apparatus according to claim 5, wherein the computing unit is specifically configured to:
the predicted compensated displacement of the sending vehicle is equal to the product of the predicted delayed speed and the predicted delay period.
9. A networked vehicle motion state information delay compensation system is characterized by comprising:
a memory for storing a computer program;
a processor for implementing the steps of the networked vehicle motion state information delay compensation method according to any one of claims 1 to 4 when executing the computer program.
10. A computer-readable medium, wherein a computer program is stored on the computer-readable medium, and when the computer program is executed, the computer program implements the steps of the networked vehicle motion state information delay compensation method according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210668381.5A CN115086375A (en) | 2022-06-14 | 2022-06-14 | Method, device, system and medium for compensating motion state information delay of networked vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210668381.5A CN115086375A (en) | 2022-06-14 | 2022-06-14 | Method, device, system and medium for compensating motion state information delay of networked vehicle |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115086375A true CN115086375A (en) | 2022-09-20 |
Family
ID=83252174
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210668381.5A Pending CN115086375A (en) | 2022-06-14 | 2022-06-14 | Method, device, system and medium for compensating motion state information delay of networked vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115086375A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117395691A (en) * | 2023-12-11 | 2024-01-12 | 泉州市三川通讯技术股份有限责任公司 | Vehicle-mounted terminal communication optimization method under weak signal environment |
-
2022
- 2022-06-14 CN CN202210668381.5A patent/CN115086375A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117395691A (en) * | 2023-12-11 | 2024-01-12 | 泉州市三川通讯技术股份有限责任公司 | Vehicle-mounted terminal communication optimization method under weak signal environment |
| CN117395691B (en) * | 2023-12-11 | 2024-03-01 | 泉州市三川通讯技术股份有限责任公司 | Vehicle-mounted terminal communication optimization method under weak signal environment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11799936B2 (en) | Low latency wireless communication system for teleoperated vehicle environments | |
| Mei et al. | Realtime mobile bandwidth prediction using lstm neural network | |
| WO2022156182A1 (en) | Methods and apparatuses for constructing vehicle dynamics model and for predicting vehicle state information | |
| CN112037518B (en) | Calculation method and system for predicting traffic state and vehicle | |
| US11665017B2 (en) | Telemetry reporting in vehicle super resolution systems | |
| US20190266498A1 (en) | Behavioral models for vehicles | |
| CN113561994A (en) | Trajectory planning method and device, storage medium and electronic equipment | |
| CN115086375A (en) | Method, device, system and medium for compensating motion state information delay of networked vehicle | |
| CN113382383A (en) | Method for unloading calculation tasks of public transport vehicle based on strategy gradient | |
| Aparna et al. | Steering angle prediction for autonomous driving using federated learning: The impact of vehicle-to-everything communication | |
| CN113643532A (en) | Regional traffic prediction method and device | |
| Li et al. | Predictive control for vehicular sensor networks based on round-trip time-delay prediction | |
| CN112612277B (en) | Vehicle trajectory planning method, device, equipment and computer readable medium | |
| CN112373471B (en) | Method, device, electronic equipment and readable medium for controlling vehicle running | |
| US20230121913A1 (en) | Intelligent messaging framework for vehicle ecosystem communication | |
| CN114169463A (en) | Autonomous prediction lane information model training method and device | |
| Zanma et al. | Optimal control input for discrete‐time networked control systems with data dropout | |
| CN113421444A (en) | Urban road network signal control method and device based on vehicle path information | |
| CN112636719A (en) | ILC system input signal filtering method under data loss and channel noise interference | |
| Hegedűs et al. | Implementation and design of ultra-local model-based control strategy for autonomous vehicles | |
| CN113593241B (en) | Vehicle interaction information verification method and device, electronic equipment and readable medium | |
| CN116009549B (en) | Vehicle queue dynamic reorganization method and device for non-networked tandem vehicles | |
| WO2023206532A1 (en) | Prediction method and apparatus, electronic device and computer-readable storage medium | |
| US20220237433A1 (en) | Data driven recognition of anomalies and continuation of sensor data | |
| CN116009392B (en) | Quantizer-based asynchronous event trigger control method, quantizer-based asynchronous event trigger control device, quantizer-based asynchronous event trigger control equipment and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |