CN117198065A - Intelligent speed limiter for automobile - Google Patents
Intelligent speed limiter for automobile Download PDFInfo
- Publication number
- CN117198065A CN117198065A CN202311302720.9A CN202311302720A CN117198065A CN 117198065 A CN117198065 A CN 117198065A CN 202311302720 A CN202311302720 A CN 202311302720A CN 117198065 A CN117198065 A CN 117198065A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- layer
- node
- nodes
- graph structure
- 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.)
- Granted
Links
- 239000013598 vector Substances 0.000 claims abstract description 52
- 238000000547 structure data Methods 0.000 claims abstract description 42
- 238000002910 structure generation Methods 0.000 claims abstract description 5
- 238000000034 method Methods 0.000 claims description 12
- 230000004913 activation Effects 0.000 claims description 6
- 230000002776 aggregation Effects 0.000 claims description 6
- 238000004220 aggregation Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention relates to the technical field of vehicle control, and discloses an intelligent speed limiter of an automobile, which comprises the following components: the path information acquisition module is used for defining a generation area with a radius being a first radius according to the position of the first vehicle at the t time point; collecting position information of a vehicle in a generation area and path information of a past time period with the length of C; the graph structure generation module constructs graph structure data based on the collected information, wherein the graph structure data comprises nodes and initial vectors of the nodes; the image acquisition module acquires an image right in front of the first vehicle at a t time point as a first image; the speed limit value generation module inputs the graph structure data and the first image into the first model and outputs a speed limit value; the speed limiting control module limits the first vehicle to exceed a speed limiting value; the invention can adapt to overtaking situations to provide speed limit.
Description
Technical Field
The invention relates to the technical field of vehicle control, in particular to an intelligent speed limiter for an automobile.
Background
The general speed limit is limited according to the speed of the front vehicle and the safety distance to be maintained, but when other vehicles overtake the current vehicle, the overtaking vehicle can be used as a new front vehicle of the current vehicle, and the overtaking vehicle can not maintain the safety distance with the overtaking vehicle when overtaking at the current speed, so that the speed limit method according to the speed of the front vehicle and the safety distance to be maintained is not applicable any more.
Disclosure of Invention
The invention provides an intelligent speed limiter for an automobile, which solves the technical problem that the speed limiter cannot adapt to overtaking situations to provide speed limit in the related technology.
The invention provides an intelligent speed limiter of an automobile, which comprises the following components:
the path information acquisition module is used for defining a generation area with the radius being the first radius according to the position of the first vehicle at a t time point, and defining vehicles except the first vehicle in the generation area as second vehicles;
collecting position information of a vehicle in a generation area and path information of a past time period with the length of C;
the graph structure generation module constructs graph structure data based on the collected information, wherein the graph structure data comprises nodes and initial vectors of the nodes;
nodes of the graph structure data are in one-to-one correspondence with vehicles in the generation area;
the neighbor node set of the ith node of the graph structure data is N i M is more than or equal to i is more than or equal to 1, M is the total number of nodes, N i The nodes in the tree are all provided with edges with the ith node, N i The vehicle corresponding to the node in (a) is positioned in a first area taking the ith node as the center at a time point t, and the radius of the first area is a second radius;
the image acquisition module acquires an image right in front of the first vehicle at a t time point as a first image;
the speed limit value generation module inputs the graph structure data and the first image into a first model, wherein the first model comprises a GNN layer, a convolution layer, a splicing layer and a full connection layer, the GNN layer inputs the graph structure data, an embedded vector of a node is output, the convolution layer inputs the first image, a first image characteristic is output, the splicing layer splices the embedded vector of the node corresponding to the first vehicle with the first image characteristic and then inputs the full connection layer, and the full connection layer outputs the speed limit value of the first vehicle in a time period with the length D after a time point t;
and the speed limiting control module is used for detecting whether the speed of the first vehicle exceeds a speed limiting value, decelerating the first vehicle if the speed exceeds the speed limiting value, and limiting the speed to exceed the speed limiting value.
Further, the convolutional layer is pre-trained by a single full-join layer that joins the output visibility values.
Further, the speed limit control module is connected with a controller of the power source of the first vehicle and sends a speed limit signal to the controller.
Further, the path information of a past one time period of length C of the vehicle in the generation area is acquired by the in-vehicle sensor, and the path information is expressed as:wherein->Feature vectors representing the t-C time point to the t time point of the ith vehicle, respectively, +.> Wherein->And->The longitude and latitude of the ith time point of the ith vehicle are respectively indicated.
Further, the initial vector of the nodes of the graph structure data is represented by vectorizing the path information of the past one time period of length C of the vehicle.
Further, the calculation formula of the GNN layer is as follows:
wherein O is i Node update vector, θ, representing the i-th node i Aggregation index, N, representing the ith node i Representing a set of nodes having edges with the ith node, σ represents a nonlinear activation function, W 2 Representing a second weight, E j Respectively represent the j-th nodeA node vector;
the calculation formula of the aggregation index of the ith node is as follows:
wherein Z is i ＝W 1 E i ，Z j ＝W 1 E j ，E i And E is j Node vectors representing the ith and j-th nodes, W 1 A first weight is indicated and a second weight is indicated,representing a third weight, T representing a transpose operation, exp representing a natural exponential function, leakyReLU representing a LeakyRelu activation function, N i Representing a collection of nodes that have edges with the ith node.
Further, independent training is carried out on the GNN layer, the convolution layer and the splicing layer of the first model, the second splicing layer is connected behind the splicing layer, the input of the second splicing layer is also connected with an initial characteristic generator, the second splicing layer is connected with a generator, the generator is connected with a discriminator, and the input of the discriminator is also connected with a termination characteristic generator;
marking a second vehicle overtaking the first vehicle as an overtaking vehicle in a time period with the length D after the t time point;
the initial feature generator generates initial features according to the path information of the overtaking vehicle in a time period with the length of C, which is passed by the overtaking vehicle at the t time point;
the stop feature generator generates a stop feature according to the position information of the overtaking vehicle when overtaking to the same lane of the first vehicle in the time period with the length D after the t time point;
the second splicing layer is used for splicing the splicing characteristics output by the splicing layer and the initial characteristics to form a combined vector, the combined vector is input to the generator, the generator outputs a generated vector, and the dimension of the generated vector is consistent with the dimension of the termination characteristics.
Further, the generator trains the loss function as:
L＝1*logD(x 1 )+0*logD(x 0 )+1*logD(G(z 1 ))+0*logD(G(z 0 ))
wherein L represents a loss value, D (x 0 )、(x 1 ) Judging whether the characteristics of the input discriminators are the generation termination characteristics or the probability of generating vectors by the discriminators respectively; wherein D (G (z) 1 ))、D(G(z 0 ) A probability that the arbiter determines whether the sample input to the arbiter is a generated termination feature or a generated vector, respectively.
The invention provides an intelligent speed limiting method for an automobile, which comprises the following steps of:
s201, defining a generation area with a radius being a first radius by using the position of a first vehicle at a t time point, and defining vehicles except the first vehicle in the generation area as second vehicles at the t time point;
s202, collecting position information of a vehicle in a generation area and path information of a past time period with the length of C;
s203, constructing graph structure data based on the acquired information, wherein the graph structure data comprises nodes and initial vectors of the nodes;
nodes of the graph structure data are in one-to-one correspondence with vehicles in the generation area;
the neighbor node set of the ith node of the graph structure data is N i M is more than or equal to i is more than or equal to 1, M is the total number of nodes, N i The nodes in the tree are all provided with edges with the ith node, N i The vehicle corresponding to the node in (a) is positioned in a first area taking the ith node as the center at a time point t, and the radius of the first area is a second radius;
s204, acquiring an image right in front of a first vehicle at a t time point as a first image;
s205, inputting the graph structure data and a first image into a first model, wherein the first model comprises a GNN layer, a convolution layer, a splicing layer and a full connection layer, the GNN layer inputs the graph structure data, an embedded vector of a node is output, the convolution layer inputs the first image, a first image feature is output, the splicing layer splices the embedded vector of the node corresponding to a first vehicle with the first image feature and then inputs the full connection layer, and the full connection layer outputs a speed limit value of the first vehicle in a time period with the length D after a time point t;
s206, detecting whether the speed of the first vehicle exceeds the speed limit value, if so, decelerating the first vehicle and limiting the speed to exceed the speed limit value.
The present invention provides a storage medium storing non-transitory computer readable instructions that, when executed by a computer, perform one or more of the steps of the foregoing intelligent speed limiting method for a vehicle.
The invention can adapt to overtaking situations to provide speed limit and output proper speed limit speed under different visibility conditions.
Drawings
FIG. 1 is a schematic block diagram of an intelligent speed limiter for an automobile in accordance with the present invention;
FIG. 2 is a flow chart of an intelligent speed limiting method for an automobile according to the present invention;
FIG. 3 is a block diagram of a storage medium of the present invention;
fig. 4 is a graph of the output results of the first model of the present invention.
In the figure: the system comprises a path information acquisition module 101, a graph structure generation module 102, an image acquisition module 103, a speed limit value generation module 104 and a speed limit control module 105.
Detailed Description
The subject matter described herein will now be discussed with reference to example embodiments. It is to be understood that these embodiments are merely discussed so that those skilled in the art may better understand and implement the subject matter described herein and that changes may be made in the function and arrangement of the elements discussed without departing from the scope of the disclosure herein. Various examples may omit, replace, or add various procedures or components as desired. In addition, features described with respect to some examples may be combined in other examples as well.
Fig. 1 is a schematic diagram of an intelligent speed limiter for an automobile according to at least one embodiment of the present disclosure, including:
a path information acquisition module 101, configured to define a generation area with a radius being a first radius by a position of a first vehicle at a t time point, and define vehicles other than the first vehicle within the generation area as second vehicles;
collecting position information of a vehicle in a generation area and path information of a past time period with the length of C;
the graph structure generation module 102 constructs graph structure data based on the collected information, the graph structure data including nodes and initial vectors of the nodes;
nodes of the graph structure data are in one-to-one correspondence with vehicles in the generation area;
the neighbor node set of the ith node of the graph structure data is N i M is more than or equal to i is more than or equal to 1, M is the total number of nodes, N i The nodes in the tree are all provided with edges with the ith node, N i The vehicle corresponding to the node in (a) is positioned in a first area taking the ith node as the center at a time point t, and the radius of the first area is a second radius;
an image acquisition module 103 that acquires an image immediately in front of the first vehicle at a time point t as a first image;
the speed limit value generating module 104 inputs the graph structure data and the first image into a first model, wherein the first model comprises a GNN layer, a convolution layer, a splicing layer and a full connection layer, the GNN layer inputs the graph structure data, an embedded vector of a node is output, the convolution layer inputs the first image, a first image feature is output, the splicing layer splices the embedded vector of the node corresponding to the first vehicle with the first image feature and then inputs the full connection layer, and the full connection layer outputs the speed limit value of the first vehicle in a time period with the length D after a time point t;
the speed limit control module 105 detects whether the speed of the first vehicle exceeds a speed limit value, and if so, decelerates the first vehicle and limits its speed to exceed the speed limit value.
The first vehicle is the vehicle which needs to limit speed.
In one embodiment of the present disclosure, the convolutional layer is pre-trained by a fully connected layer that individually connects output visibility values;
in one embodiment of the present disclosure, the first image may be provided by a vehicle recorder of the first vehicle.
In one embodiment of the present disclosure, the speed limit control module 105 is coupled to a controller of the power source of the first vehicle to which the speed limit signal is sent.
In one embodiment of the present disclosure, collecting, by a car navigation system, position information of a vehicle within a generation area, the position information including: longitude and latitude of the vehicle; the time intervals of the acquisitions are equal.
In one embodiment of the present disclosure, path information of a past one time period of length C of a vehicle in a generation area is acquired by an in-vehicle sensor, the path information being expressed as:wherein->Feature vectors representing the t-C time point to the t time point of the ith vehicle, respectively, +.>Wherein->And->Longitude and latitude of the t time point of the i-th vehicle are respectively shown;
in one embodiment of the present disclosure, the initial vector of nodes of the graph structure data is represented by vectorizing path information of a past one time period of length C of the vehicle;
the running record, which is intercepted during the first model training and always keeps a safe distance with the front vehicle, generates a training sample, wherein the running record of the generated training sample contains the condition of overtaking the first vehicle, and the first vehicle keeps a set safe distance with the front vehicle in the whole running record. The travel record includes a t time point, the travel record before the t time point is a travel front stage, the travel record after the t time point is a travel rear stage, the highest speed of the first vehicle in the travel rear stage is used as a speed limit value corresponding to a training sample, and the highest speed is compared with the speed limit value output by the first model to be used as a loss.
The value range of the value of D is preferably 1-10s, at this time, the probability of the erroneous speed limit value output by the first model is very low, and a larger value of D can accompany overtaking for more times in the process, so that the complexity is improved, and the probability of the erroneous speed limit value output by the first model is increased.
The value of C is preferably between 20 and 50 seconds, too large a value of C will result in too high a processing speed, and too small a value of C will result in too little information being produced.
In one embodiment of the present disclosure, the set safe distance is selected according to the visibility corresponding to the travel record.
In one embodiment of the present disclosure, the calculation formula of the GNN layer is as follows:
wherein O is i Node update vector, θ, representing the i-th node i Aggregation index, N, representing the ith node i Representing a set of nodes having edges with the ith node, σ represents a nonlinear activation function, W 2 Representing a second weight, E j Node vectors respectively representing jth nodes;
the calculation formula of the aggregation index of the ith node is as follows:
wherein Z is i ＝W 1 E i ，Z j ＝W 1 E j ，E i And E is j Node vectors representing the ith and j-th nodes, W 1 A first weight is indicated and a second weight is indicated,representing a third weight, T representing a transpose operation, exp representing a natural exponential function, leakyReLU representing a LeakyRelu activation function, N i A set of nodes representing edges that exist with the ith node;
fig. 2 is a schematic diagram of an automotive speed limiting method according to at least one embodiment of the present disclosure, including the steps of:
s201, defining a generation area with a radius being a first radius by using the position of a first vehicle at a t time point, and defining vehicles except the first vehicle in the generation area as second vehicles at the t time point;
s202, collecting position information of a vehicle in a generation area and path information of a past time period with the length of C;
s203, constructing graph structure data based on the acquired information, wherein the graph structure data comprises nodes and initial vectors of the nodes;
nodes of the graph structure data are in one-to-one correspondence with vehicles in the generation area;
the neighbor node set of the ith node of the graph structure data is N i M is more than or equal to i is more than or equal to 1, M is the total number of nodes, N i The nodes in the tree are all provided with edges with the ith node, N i The vehicle corresponding to the node in (a) is positioned in a first area taking the ith node as the center at a time point t, and the radius of the first area is a second radius;
s204, acquiring an image right in front of a first vehicle at a t time point as a first image;
s205, inputting the graph structure data and a first image into a first model, wherein the first model comprises a GNN layer, a convolution layer, a splicing layer and a full connection layer, the GNN layer inputs the graph structure data, an embedded vector of a node is output, the convolution layer inputs the first image, a first image feature is output, the splicing layer splices the embedded vector of the node corresponding to a first vehicle with the first image feature and then inputs the full connection layer, and the full connection layer outputs a speed limit value of the first vehicle in a time period with the length D after a time point t;
s206, detecting whether the speed of the first vehicle exceeds the speed limit value, if so, decelerating the first vehicle and limiting the speed to exceed the speed limit value.
Fig. 3 is a storage medium 300 storing non-transitory computer readable instructions 310 according to at least one embodiment of the present disclosure, where the non-transitory computer readable instructions 310 are used by a computer to perform one or more steps of the intelligent speed limiting method of a vehicle
The model is based on the cut-out history overtaking segment capable of keeping the safe distance as a training sample, and the training sample contains the condition that the first vehicle can keep the safe distance after the speed limit is improved, so that the training sample is difficult to correct manually, and the speed limit value output by the model is lower than the maximum speed value which can be adopted by the first vehicle under most conditions.
In one embodiment of the present disclosure, the following method is provided: after the training process is finished, independent training is carried out on the GNN layer, the convolution layer and the splicing layer of the first model, the second splicing layer is connected after the splicing layer, the input of the second splicing layer is also connected with an initial characteristic generator, the second splicing layer is connected with a generator, the generator is connected with a discriminator, and the input of the discriminator is also connected with a termination characteristic generator;
marking a second vehicle overtaking the first vehicle as an overtaking vehicle in a time period with the length D after the t time point;
the initial feature generator generates initial features according to the path information of the overtaking vehicle in a time period with the length of C, which is passed by the overtaking vehicle at the t time point;
the stop feature generator generates a stop feature according to the position information of the overtaking vehicle when overtaking to the same lane of the first vehicle in the time period with the length D after the t time point;
the second splicing layer is used for splicing the splicing characteristics output by the splicing layer and the initial characteristics to form a combined vector, inputting the combined vector into the generator, outputting a generated vector by the generator, and enabling the dimension of the generated vector to be consistent with the dimension of the termination characteristics;
because the number of overtaking vehicles is not fixed, the number of position information of overtaking vehicles is also different, the dimension of the termination feature generated by the same rule code is different, and the dimension of the termination feature is required to be the same as the dimension of the initial feature by interpolation of a 0 value;
the trained loss function is:
L＝1*logD(x 1 )+0*logD(x 0 )+1*logD(G(z 1 ))+0*logD(G(z 0 ))
wherein L represents a loss value, D (x 0 )、(x 1 ) Judging whether the characteristics of the input discriminators are the generation termination characteristics or the probability of generating vectors by the discriminators respectively; wherein D (G (z) 1 ))、D(G(z 0 ) The probability that the arbiter judges whether the sample input to the arbiter generates the termination feature or the vector, respectively;
the loss function is added with a negative sign when training the arbiter.
The first model is used for extracting characteristics related to the speed of a first vehicle which is at a safe distance when other vehicles overtake in the next time period, the extraction capacity of the first model for the required characteristics can be optimized through training of generating the overtaking position of the overtaking vehicle in training, and the output speed limit value of the first model is approximate to the maximum speed value which can be adopted by the first vehicle;
as shown in fig. 4, a group of training samples is randomly selected to test by adding the first models before and after training, a higher speed limit value is output by adding the first models after training to most of the samples, and the situation that the first vehicle and the front vehicle are smaller than a safe distance does not occur after the speed limit value output by the first models before and after training is combined with the driving record of the source of the training samples to perform the simulation test.
The embodiment has been described above with reference to the embodiment, but the embodiment is not limited to the above-described specific implementation, which is only illustrative and not restrictive, and many forms can be made by those of ordinary skill in the art, given the benefit of this disclosure, are within the scope of this embodiment.
Claims (10)
1. An intelligent speed limiter for an automobile, comprising:
the path information acquisition module is used for defining a generation area with the radius being the first radius according to the position of the first vehicle at a t time point, and defining vehicles except the first vehicle in the generation area as second vehicles;
collecting position information of a vehicle in a generation area and path information of a past time period with the length of C;
the graph structure generation module constructs graph structure data based on the collected information, wherein the graph structure data comprises nodes and initial vectors of the nodes;
nodes of the graph structure data are in one-to-one correspondence with vehicles in the generation area;
the neighbor node set of the ith node of the graph structure data is N i M is more than or equal to i is more than or equal to 1, M is the total number of nodes, N i The nodes in the tree are all provided with edges with the ith node, N i The vehicle corresponding to the node in (a) is positioned in a first area taking the ith node as the center at a time point t, and the radius of the first area is a second radius;
the image acquisition module acquires an image right in front of the first vehicle at a t time point as a first image;
the speed limit value generation module inputs the graph structure data and the first image into a first model, wherein the first model comprises a GNN layer, a convolution layer, a splicing layer and a full connection layer, the GNN layer inputs the graph structure data, an embedded vector of a node is output, the convolution layer inputs the first image, a first image characteristic is output, the splicing layer splices the embedded vector of the node corresponding to the first vehicle with the first image characteristic and then inputs the full connection layer, and the full connection layer outputs the speed limit value of the first vehicle in a time period with the length D after a time point t;
and the speed limiting control module is used for detecting whether the speed of the first vehicle exceeds a speed limiting value, decelerating the first vehicle if the speed exceeds the speed limiting value, and limiting the speed to exceed the speed limiting value.
2. An intelligent speed limiter in accordance with claim 1 wherein the convolutional layer is pre-trained by a single full link layer that links the output visibility values.
3. An intelligent speed limiter in a vehicle as claimed in claim 1 wherein the speed limit control module is connected to the controller of the power source of the first vehicle and sends a speed limit signal thereto.
4. The intelligent speed limiter for vehicle according to claim 1, wherein the path information of a past time period of length C of the vehicle in the generation area is collected by the in-vehicle sensor, and the path information is expressed as:wherein->Feature vectors representing the t-C time point to the t time point of the ith vehicle, respectively, +.>Wherein->And->The longitude and latitude of the ith time point of the ith vehicle are respectively indicated.
5. An intelligent speed limiter for vehicles according to claim 1 wherein the initial vector of nodes of the graph structure data is represented by vectorizing the path information of the vehicle over a period of time of length C.
6. The intelligent speed limiter for the automobile according to claim 1, wherein the GNN layer has the following formula:
wherein O is i Node update vector, θ, representing the i-th node i Aggregation index, N, representing the ith node i Representing a set of nodes having edges with the ith node, σ represents a nonlinear activation function, W 2 Representing a second weight, E j Node vectors respectively representing jth nodes;
the calculation formula of the aggregation index of the ith node is as follows:
wherein Z is i ＝W 1 E i ，Z j ＝W 1 E j ，E i And E is j Node vectors representing the ith and j-th nodes, W 1 A first weight is indicated and a second weight is indicated,representing a third weight, T representing a transpose operation, exp representing a natural exponential function, leakyReLU representing a LeakyRelu activation function, N i Representing a collection of nodes that have edges with the ith node.
7. The intelligent speed limiter for the automobile according to claim 1, wherein after training the first model, independent training is carried out on a GNN layer, a convolution layer and a splicing layer of the first model, the splicing layer is connected with a second splicing layer, the input of the second splicing layer is also connected with a starting characteristic generator, the second splicing layer is connected with a generator, the generator is connected with a discriminator, and the input of the discriminator is also connected with a termination characteristic generator;
marking a second vehicle overtaking the first vehicle as an overtaking vehicle in a time period with the length D after the t time point;
the initial feature generator generates initial features according to the path information of the overtaking vehicle in a time period with the length of C, which is passed by the overtaking vehicle at the t time point;
the stop feature generator generates a stop feature according to the position information of the overtaking vehicle when overtaking to the same lane of the first vehicle in the time period with the length D after the t time point;
the second splicing layer is used for splicing the splicing characteristics output by the splicing layer and the initial characteristics to form a combined vector, the combined vector is input to the generator, the generator outputs a generated vector, and the dimension of the generated vector is consistent with the dimension of the termination characteristics.
8. The intelligent speed limiter according to claim 7 wherein the generator trains the loss function as:
L＝1*logD(x 1 )+0*logD(x 0 )+1*logD(G(z 1 ))+0*logD(G(z 0 ))
wherein L represents a loss value, D (x 0 )、(x 1 ) Judging whether the characteristics of the input discriminators are the generation termination characteristics or the probability of generating vectors by the discriminators respectively; wherein D (G (z) 1 ))、D(G(z 0 ) A probability that the arbiter determines whether the sample input to the arbiter is a generated termination feature or a generated vector, respectively.
9. An intelligent speed limiting method for an automobile is characterized by comprising the following steps of:
s201, defining a generation area with a radius being a first radius by using the position of a first vehicle at a t time point, and defining vehicles except the first vehicle in the generation area as second vehicles at the t time point;
s202, collecting position information of a vehicle in a generation area and path information of a past time period with the length of C;
s203, constructing graph structure data based on the acquired information, wherein the graph structure data comprises nodes and initial vectors of the nodes;
nodes of the graph structure data are in one-to-one correspondence with vehicles in the generation area;
the neighbor node set of the ith node of the graph structure data is N i M is more than or equal to i is more than or equal to 1, M is the total number of nodes, N i The nodes in the tree are all provided with edges with the ith node, N i The vehicle corresponding to the node in the vehicle is positioned in the ith node at the t time pointA first region of the core having a radius of a second radius;
s204, acquiring an image right in front of a first vehicle at a t time point as a first image;
s205, inputting the graph structure data and a first image into a first model, wherein the first model comprises a GNN layer, a convolution layer, a splicing layer and a full connection layer, the GNN layer inputs the graph structure data, an embedded vector of a node is output, the convolution layer inputs the first image, a first image feature is output, the splicing layer splices the embedded vector of the node corresponding to a first vehicle with the first image feature and then inputs the full connection layer, and the full connection layer outputs a speed limit value of the first vehicle in a time period with the length D after a time point t;
s206, detecting whether the speed of the first vehicle exceeds the speed limit value, if so, decelerating the first vehicle and limiting the speed to exceed the speed limit value.
10. A storage medium having stored thereon non-transitory computer readable instructions which, when executed by a computer, perform one or more steps of the intelligent speed limiting method of a vehicle as claimed in claim 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311302720.9A CN117198065B (en) | 2023-10-09 | 2023-10-09 | Intelligent speed limiter for automobile |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311302720.9A CN117198065B (en) | 2023-10-09 | 2023-10-09 | Intelligent speed limiter for automobile |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117198065A true CN117198065A (en) | 2023-12-08 |
| CN117198065B CN117198065B (en) | 2024-05-10 |
Family
ID=88999905
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311302720.9A Active CN117198065B (en) | 2023-10-09 | 2023-10-09 | Intelligent speed limiter for automobile |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117198065B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103359123A (en) * | 2013-07-04 | 2013-10-23 | 陈根 | Intelligent vehicle speed control and management system and implementing method thereof |
| CN105984341A (en) * | 2015-02-15 | 2016-10-05 | 广州汽车集团股份有限公司 | Automobile driving speed limiting method and system |
| DE102019114595A1 (en) * | 2018-08-07 | 2020-02-13 | GM Global Technology Operations LLC | INTELLIGENT VEHICLE NAVIGATION SYSTEMS, METHOD AND CONTROL LOGIC FOR DERIVING ROAD SECTION SPEED LIMITS |
| CN113147733A (en) * | 2021-04-30 | 2021-07-23 | 东风汽车集团股份有限公司 | Intelligent speed limiting system and method for automobile in rain, fog and sand-dust weather |
| CN114299742A (en) * | 2022-01-20 | 2022-04-08 | 福建工程学院 | Dynamic recognition and updating recommendation method for speed limit information of expressway |
| CN114822055A (en) * | 2022-06-06 | 2022-07-29 | 深圳英博达智能科技有限公司 | Intelligent traffic road cooperation system based on machine vision detection |
| DE102021109716A1 (en) * | 2021-04-16 | 2022-10-20 | Bayerische Motoren Werke Aktiengesellschaft | Speed limiter for a vehicle |
-
2023
- 2023-10-09 CN CN202311302720.9A patent/CN117198065B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103359123A (en) * | 2013-07-04 | 2013-10-23 | 陈根 | Intelligent vehicle speed control and management system and implementing method thereof |
| CN105984341A (en) * | 2015-02-15 | 2016-10-05 | 广州汽车集团股份有限公司 | Automobile driving speed limiting method and system |
| DE102019114595A1 (en) * | 2018-08-07 | 2020-02-13 | GM Global Technology Operations LLC | INTELLIGENT VEHICLE NAVIGATION SYSTEMS, METHOD AND CONTROL LOGIC FOR DERIVING ROAD SECTION SPEED LIMITS |
| DE102021109716A1 (en) * | 2021-04-16 | 2022-10-20 | Bayerische Motoren Werke Aktiengesellschaft | Speed limiter for a vehicle |
| CN113147733A (en) * | 2021-04-30 | 2021-07-23 | 东风汽车集团股份有限公司 | Intelligent speed limiting system and method for automobile in rain, fog and sand-dust weather |
| CN114299742A (en) * | 2022-01-20 | 2022-04-08 | 福建工程学院 | Dynamic recognition and updating recommendation method for speed limit information of expressway |
| CN114822055A (en) * | 2022-06-06 | 2022-07-29 | 深圳英博达智能科技有限公司 | Intelligent traffic road cooperation system based on machine vision detection |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117198065B (en) | 2024-05-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2022083784A1 (en) | Road detection method based on internet of vehicles | |
| CN107169567B (en) | Method and device for generating decision network model for automatic vehicle driving | |
| CN108108766A (en) | Driving behavior recognition methods and system based on Fusion | |
| CN112307978B (en) | Target detection method and device, electronic equipment and readable storage medium | |
| WO2022027894A1 (en) | Driver behavior detection method and apparatus, electronic device, storage medium and program | |
| CN111338385A (en) | A vehicle following method based on the fusion of GRU network model and Gipps model | |
| CN112990065A (en) | Optimized YOLOv5 model-based vehicle classification detection method | |
| CN110298374A (en) | A kind of driving locus energy consumption analysis method and apparatus based on deep learning | |
| CN111507488A (en) | VR-based vehicle maintenance auxiliary system | |
| US20230394849A1 (en) | Methods and apparatus for automatic collection of under-represented data for improving a training of a machine learning model | |
| CN112465031B (en) | Data classification method, device and computer readable storage medium | |
| CN110570867A (en) | Voice processing method and system for locally added corpus | |
| CN110728459A (en) | Travel mode identification system, method and device and model training method and device | |
| CN117649906A (en) | Casting quality prediction method for integrated aluminum alloy structural part, electronic equipment and medium | |
| CN112380918A (en) | Road vehicle state identification method and device, electronic equipment and storage medium | |
| CN116894394A (en) | Automatic driving test scene generation method and related equipment | |
| CN115935642A (en) | Method and system for automatic generation of extreme test scenarios for intelligent vehicles based on accident information | |
| Kang et al. | A transfer learning based abnormal can bus message detection system | |
| CN117198065B (en) | Intelligent speed limiter for automobile | |
| CN115586763A (en) | Unmanned vehicle keeps away barrier test equipment | |
| CN118314424B (en) | Vehicle-road collaborative self-advancing learning multi-mode verification method based on edge scene | |
| CN112348039B (en) | Training method of driving behavior analysis model, driving behavior analysis method and equipment | |
| CN112052829B (en) | Pilot behavior monitoring method based on deep learning | |
| CN115841712B (en) | Driving data processing method, device and equipment based on V2X technology | |
| CN115171389B (en) | Method for identifying other vehicles’ overtaking and lane-changing intentions on highways based on GMM-HMM |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |