A data communication method, platform and medium for simulation system

Abstract

The present invention relates to the field of digital information transmission technology, and specifically to a data communication method, platform and medium for a simulation system, including: obtaining a plurality of environmental parameter sequences and a plurality of model parameter sequences, obtaining an environmental complexity coefficient sequence according to the environmental parameters at each moment in the environmental parameter sequence; determining the rationality index of each model parameter corresponding to the robot model to be simulated at the current moment according to each model parameter sequence and the environmental complexity coefficient sequence; performing interactive transmission of adjustment information between the simulation end and the model end according to the rationality index, iteratively adjusting each environmental parameter, and repeatedly executing the simulation communication steps until the simulation end condition is met. The present invention realizes real-time interactive transmission of adjustment information by adaptively determining the rationality index of each model parameter in the simulation system at the current moment, thereby improving the effectiveness of data communication in the simulation system.

Description

Technical Field

[0001] The present invention relates to the technical field of digital information transmission, and in particular to a data communication method, platform and medium of a simulation system. Background Art

[0002] Multi-mode physical simulation is an advanced simulation technology that is mainly used to simulate and analyze multiple physical phenomena and interactions at the same time. It is often used in the research and verification of complex systems, such as advanced robotics and energy systems. Multi-mode physical simulation can consider multiple physical effects (such as mechanics, thermals, fluid dynamics, etc.) at the same time and interact in real time in the simulation. During the simulation interaction process, in order to verify the simulation results of the robot in different physical environments, a comprehensive simulation system is usually used to combine software and hardware resources to realize the multi-modal environment testing of the robot, including environmental modeling, integration of robot models, simulation interaction, and corresponding data recording and analysis.

[0003] In order to optimize the design of robots, reduce the risks and costs of field tests, and accelerate the development cycle of robot products, it is necessary to analyze the robots through multi-mode physical simulation, that is, to integrate the robot model into the simulation system. In the process of multi-mode physical simulation of robots, it is necessary to analyze the performance of the robots in combination with the actual complex physical environment. Simulation in complex physical environments requires a large amount of computing resources, including processing large-scale maps, motion planning of multiple robots, sensor data processing, and the movement of robots in crowds. A large amount of computing resources will occupy the bandwidth during the real-time interaction between the robot model and the simulation end, which is not conducive to the real-time interaction of important simulation information, and thus leads to low effectiveness of data communication in the simulation system. Summary of the invention

[0004] In order to solve the technical problem of low effectiveness of data communication in the above-mentioned existing simulation system, the purpose of the present invention is to provide a data communication method, platform and medium of the simulation system, and the technical scheme adopted is as follows:

[0005] An embodiment of the present invention provides a data communication method for a simulation system, the method comprising the following steps:

[0006] S1: Acquire several environmental parameter sequences corresponding to the multi-mode physical simulation environment in the current period, and several model parameter sequences corresponding to the robot model to be simulated integrated in the multi-mode physical simulation environment;

[0007] S2: According to the environmental parameters at each moment in each environmental parameter sequence, determine the environmental complexity coefficient at each moment in the current period to form an environmental complexity coefficient sequence; according to each model parameter sequence and the environmental complexity coefficient sequence, determine the rationality index of each model parameter corresponding to the robot model to be simulated at the current moment;

[0008] S3: performing interactive transmission of adjustment information between the simulation end and the model end according to the rationality index;

[0009] S4: Iteratively adjust each environmental parameter and repeat steps S1 to S3 until the simulation end condition is met.

[0010] Furthermore, the determining of the environmental complexity coefficient at each moment in the current period according to the environmental parameters at each moment in each environmental parameter sequence includes:

[0011] The environmental parameters include the density of simulated personnel, the distance between any two simulated personnel, and the moving direction of each simulated personnel; any moment is taken as the target moment, and the environmental complexity coefficient at the target moment is determined according to the density of simulated personnel, the distance between any two simulated personnel, and the moving direction of each simulated personnel at the target moment.

[0012] Furthermore, determining the environmental complexity coefficient at the target time according to the simulated personnel density at the target time, the distance between any two simulated personnel, and the moving direction of each simulated personnel includes:

[0013] The size of the total simulation area occupied by all simulated personnel at the target time is taken as the simulated personnel density; all negative correlation values ​​of the distances are accumulated to obtain the result value after the accumulation processing; and the angle between the moving direction of the simulated personnel at the target time and the preset normal moving direction of the robot is determined;

[0014] The environmental complexity coefficient at the target moment is determined by combining the simulated personnel density at the target moment, the distribution fluctuation of the angle, and the result value after the accumulation processing; wherein the simulated personnel density, the distribution fluctuation of the angle, and the result value are all positively correlated with the environmental complexity coefficient.

[0015] Furthermore, the determination of the rationality index of each model parameter corresponding to the robot model to be simulated at the current moment according to each model parameter sequence and the environmental complexity coefficient sequence includes:

[0016] Each model parameter sequence and environmental complexity coefficient sequence is subjected to curve fitting to obtain various fitting curves, the fitting curve corresponding to the environmental complexity coefficient sequence is used as the reference curve, and the fitting curve corresponding to the model parameter sequence is used as the comparison curve;

[0017] Determine the extreme value points in the reference curve and each comparison curve, form an extreme value interval with the curve intervals corresponding to two adjacent extreme value points, and obtain each extreme value interval of the reference curve and each comparison curve;

[0018] According to the slope value at each moment in each extreme value interval of the reference curve and each comparison curve, the rationality index of each model parameter corresponding to the robot model to be simulated at the current moment is determined.

[0019] Furthermore, the rationality index of each model parameter corresponding to the robot model to be simulated at the current moment is determined according to the slope value at each moment in each extreme value interval of the reference curve and each comparison curve, including:

[0020] Take any comparison curve as the target comparison curve, and take any extreme value interval with the same sequence number in the reference curve and the target comparison curve as the target extreme value interval; calculate the product of the slope at each moment in the target extreme value interval of the reference curve and the slope at the corresponding moment in the target extreme value interval of the target comparison curve, and use the sign function to process the slope product, and use the result value after data processing as the reasonable parameter determination factor at each moment;

[0021] According to the parameter reasonable determination factor and the number of moments at each moment in the target extreme value interval, the parameter reasonable determination value of the target extreme value interval of the target comparison curve is determined;

[0022] Obtain the reasonable judgment value of the parameter for each extreme value interval, calculate the cumulative value of the reasonable judgment values ​​of the parameters of all extreme value intervals of the target comparison curve, and then calculate the difference between the number of extreme value intervals of the target comparison curve and the cumulative value of the reasonable judgment value of the parameter, and determine the rationality index of the model parameter corresponding to the target comparison curve; wherein the difference is negatively correlated with the rationality index.

[0023] Further, the method of determining the reasonable determination value of the parameter of the target extreme value interval of the target comparison curve according to the reasonable determination factor of the parameter at each moment in the target extreme value interval and the number of moments includes:

[0024] Combined with the proportional relationship between the model parameters corresponding to the target comparison curve and the environmental complexity coefficient, a reasonable interval parameter judgment model is constructed; the reasonable parameter judgment factor and the number of moments at each moment in the target extreme value interval are used as input data and input into the reasonable interval parameter judgment model to obtain the reasonable parameter judgment value of the target extreme value interval of the target comparison curve.

[0025] Furthermore, the step of obtaining the reasonable interval parameter determination model includes:

[0026] When the model parameters corresponding to the target comparison curve are proportional to the environmental complexity coefficient, the expression of the interval parameter reasonable judgment model is: ; In the formula, represents the reasonable judgment value of the parameter in the uth extreme value interval, exp represents the exponential function with the natural constant as the base, represents the number of moments in the u-th extreme value interval, v represents the sequence number of moments in the u-th extreme value interval, Represents the reasonable determination factor of the parameter at the vth moment in the uth extreme value interval;

[0027] When the model parameters corresponding to the target comparison curve are inversely proportional to the environmental complexity coefficient, the expression of the interval parameter reasonable determination model is: .

[0028] Furthermore, the interactive transmission of adjustment information between the simulation end and the model end according to the rationality index includes:

[0029] Compare the rationality index of each model parameter with a preset index threshold, and select the model parameter whose rationality index is less than the preset index threshold as the model parameter to be transmitted at the simulation end;

[0030] The model parameters to be transmitted at the simulation end are sent to the model end by using a distributed transmission algorithm. The model end determines the adjustment result of each model parameter to be transmitted by using the received model parameters to be transmitted, and feeds back the adjustment result to the simulation end.

[0031] The present invention also provides a data communication platform for a simulation system, comprising a processor and a memory, wherein the processor is used for processing instructions stored in the memory to implement a data communication method for the simulation system.

[0032] The present invention also provides a computer-readable storage medium, which stores computer program codes. When the computer program codes are run on a computer, the computer executes a data communication method for a simulation system.

[0033] The present invention has the following beneficial effects:

[0034] The present invention provides a data communication method for a simulation system. The method uses a multi-mode physical simulation method to interactively communicate and transmit the real-time simulation data of a robot model between a simulation end and a model end, and quantifies the rationality of the corresponding method of the robot model by increasing the environmental complexity coefficient of the simulation environment, thereby ensuring the real-time optimization of robot model parameters. In order to overcome the defect that a large amount of computing resources required for simulation in a complex physical environment is not conducive to real-time interaction of important simulation information, the rationality index of the model parameter is determined in combination with the logical relationship between the model parameter sequence and the environmental complexity coefficient sequence, so that the rationality index can be used to preferentially transmit effective information in the subsequent transmission, thereby improving the effectiveness and efficiency of simulation data communication transmission. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0036] Figure 1 A schematic diagram of a flow chart of a data communication method of a simulation system according to an embodiment of the present invention;

[0037] Figure 2 is a schematic diagram of a robot model to be simulated in a multi-mode physical simulation environment in an embodiment of the present invention;

[0038] Figure 3 A schematic diagram of a simulation environment in an embodiment of the present invention;

[0039] Figure 4 is a flowchart of the second step of step S2 in an embodiment of the present invention;

[0040] Figure 5 is a flowchart of step S23 in an embodiment of the present invention;

[0041] Figure 6 is a flowchart of step S232 in an embodiment of the present invention;

[0042] Figure 7 4 is a flowchart of step S3 in an embodiment of the present invention. DETAILED DESCRIPTION

[0043] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the technical solutions proposed by the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0044] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0045] The application scenarios targeted by the present invention may be:

[0046] Simulation data communication refers to the data exchange communication between the simulation end and the object end in the simulation, that is, the robot model end. The purpose of communication is to perform real-time control updates on the simulation end based on the data provided by the simulation object, which can achieve the purpose and function of real-time and efficient communication. In the process of simulation testing of the robot, in order to eliminate the lack of environmental complexity, it is necessary to analyze the performance of the robot in combination with the actual complex physical environment. Simulation in a complex physical environment requires a large amount of computing resources, which will occupy the bandwidth in the real-time interaction between the robot model end and the simulation end, which is not conducive to the real-time interaction of important simulation information, thereby affecting the effectiveness of data communication in the simulation system.

[0047] In order to ensure real-time interaction between the robot model end and the simulation end during the communication process and reduce the impact of the complex physical environment in the simulation system, it is necessary to perform feedback judgment on the interaction content and give priority to the interaction of important information that is beneficial to the simulation. Specifically, this embodiment provides a data communication method for a simulation system, such as Figure 1 As shown, the following steps are included:

[0048] S1, obtaining a plurality of environmental parameter sequences corresponding to a multi-mode physical simulation environment in a current period, and a plurality of model parameter sequences corresponding to the robot model to be simulated integrated in the multi-mode physical simulation environment.

[0049] The above step S1 can be implemented by the following steps:

[0050] The first step is to build a multi-mode physical simulation environment and integrate the robot model to be simulated in the multi-mode physical simulation environment.

[0051] First of all, it should be noted that the simulation system is designed to analyze the performance results of the robot in environments of different complexity by performing multi-mode physical simulation on the robot, and to perform feedback iterative optimization of the model. Therefore, before performing performance testing and optimization of the robot model, it is necessary to build a relevant multi-mode physical simulation environment and integrate the robot model to be simulated into the constructed multi-mode physical simulation environment.

[0052] The first sub-step is to establish a mathematical model of one or more physical environments, that is, to establish a multi-mode physical simulation environment.

[0053] In a multi-mode physical simulation environment, several physical conditions that the robot model may encounter can be described, such as different ground materials, temperature, humidity and other environmental factors. The ground material can be sand, stone or grass; and it has controllable environmental parameters such as simulated personnel density, which can be used to adjust the multi-mode physical simulation environment.

[0054] Among them, the process of establishing a multi-mode physical simulation environment is a prior art and is not within the scope of protection of the present invention, and will not be elaborated on in detail here; and the above functions can all be implemented in multi-mode physical simulation, which can be specifically constructed through physical equations, empirical data and previous test results.

[0055] The second sub-step is to integrate the robot model to be simulated into the multi-mode physical simulation environment, specifically including the robot's dynamic model, control system, sensor input, etc.

[0056] In this embodiment, the robot model to be simulated needs to match the performance of the robot in the real world. First, the interactive function of the robot can specifically refer to the operation of the robot arm, the response of the sensor, etc., where the sensor-environment interaction refers to the interaction between the robot in the simulation and the environment through its virtual sensors (such as cameras, tactile sensors, distance sensors, etc.); second, the sensor will collect environmental data and feed this data back to the control system, and the control system will adjust the robot's behavior according to the sensor data, such as changing the walking path, adjusting the speed or manipulating the robot arm, and the simulation system will also simulate physical responses, such as the robot's reaction to the ground material, collision with objects, etc.

[0057] The schematic diagram of the robot model to be simulated in the multi-mode physical simulation environment is as follows: Figure 2 As shown, in Figure 2 In the simulation, the robot model to be simulated has different types of sensors, and there are related factors of controllable environmental complexity in the simulation environment, namely, crowd density, which is also the density of simulated personnel.

[0058] It is worth mentioning that through the construction of the above-mentioned simulation environment and robot integrated model, it is ensured that the simulation environment has a certain physical reality and has a controllable environmental control module, that is, controllable environmental parameters such as simulated personnel density. The environmental parameters will participate in the optimization and adjustment process in the process of optimizing the simulation communication data transmission interaction, which can be used to increase the performance of the robot model.

[0059] The second step is to obtain several environmental parameter sequences corresponding to the multi-mode physical simulation environment in the current period, and several model parameter sequences corresponding to the integration of the robot model to be simulated in the multi-mode physical simulation environment.

[0060] Specifically, each type of environmental parameter and model parameter at each moment in the current period can be collected through the simulation system, and the parameters of the same type can be grouped into a sequence, thereby obtaining each environmental parameter sequence and each model parameter sequence.

[0061] It is worth mentioning that by analyzing the changing relationship characteristics between environmental parameters and model parameters, more important data can be screened out from a large number of data to be transmitted for transmission, which reduces the amount of data transmission and, to a certain extent, overcomes the defect that complex physical environment is not conducive to real-time interaction of important simulation information, which is beneficial to enhancing the effectiveness of data communication in the simulation system.

[0062] In this embodiment, the current time period can take an empirical value of 30s. The types of environmental parameters may include the density of simulated personnel, the distance between any two simulated personnel, and the movement direction of each simulated personnel. By analyzing the changes in environmental parameters over a period of time, the environmental complexity coefficient of the simulation environment can be quantified; the types of model parameters may include the real-time speed of the robot model, the total length of the path planning, and the number of turning points.

[0063] Among them, when the angle changes more than 5° during movement, it is called an inflection point; the data collection frequency of environmental parameters and model parameters can be taken as an empirical value of once every 2 seconds. The current period, the type of model parameters, the definition of inflection points, and the frequency of data collection can all be determined by the implementer according to the actual situation. The type of model parameters is not specifically limited here.

[0064] So far, this embodiment has obtained multiple environment parameter sequences and multiple model parameter sequences corresponding to the multi-mode physical simulation environment.

[0065] S2, according to the environmental parameters at each moment in each environmental parameter sequence, determine the environmental complexity coefficient at each moment in the current period to form an environmental complexity coefficient sequence; according to each model parameter sequence and the environmental complexity coefficient sequence, determine the rationality index of each model parameter corresponding to the robot model to be simulated at the current moment.

[0066] First of all, it should be noted that when the robot model is optimized in a multi-mode physical simulation environment, it is necessary to provide real-time feedback on the adaptability of the robot model parameters at the current moment in response to complex environments. However, there are many types of model parameters. If all model parameters are fed back from the simulation end to the model end regardless of their importance, the bandwidth in the interaction process will be under great pressure, which is not conducive to the real-time interaction of important simulation information and is likely to affect the effectiveness of data communication. Therefore, in order to solve the above-mentioned technical problems, it is necessary to quantify the rationality of each model parameter of the robot model at the current moment, that is, to determine the rationality index, which refers to the rationality of the robot model in responding to different environments.

[0067] The above step S2 can be implemented by the following steps:

[0068] The first step is to determine the environmental complexity coefficient at each moment in the current period according to the environmental parameters at each moment in each environmental parameter sequence to form an environmental complexity coefficient sequence.

[0069] It should be noted that the environmental complexity coefficient can characterize the environmental complexity of the current simulation environment, and the purpose of determining the environmental complexity coefficient is to quantify the adaptability of the robot model to the simulation environment in the current environment, so as to facilitate the subsequent analysis of the relationship between the environmental complexity and the theoretical speed or path of the robot, and use the current model parameters to perform feedback updates on the robot model to ensure the purpose of real-time communication feedback and optimize the robot model parameters during the performance test.

[0070] In this embodiment, the calculation process of the environmental complexity coefficient at each moment in the current period is consistent. In order to reduce unnecessary descriptions, any moment is taken as the target moment, and the environmental complexity coefficient of the target moment is determined by taking the target moment as an example. For the multi-mode physical simulation environment and its corresponding robot model, the robot model controls its initial speed to 2m / s. The initial speed can be set by the implementer according to the specific actual situation, and the path is the normal driving direction; the control end of the multi-mode physical simulation system randomly generates several simulated personnel in the robot's driving direction at the target moment, and analyzes the environmental complexity coefficient of the target moment.

[0071] According to the simulated personnel density at the target time, the distance between any two simulated personnel, and the moving direction of each simulated personnel, the environmental complexity coefficient at the target time is determined. The specific implementation process may include:

[0072] In the first sub-step, the size of the total simulation area occupied by all simulated personnel at the target moment is taken as the simulated personnel density; the negative correlation values ​​of all distances are accumulated to obtain the result value after the accumulation processing; and the angle between the moving direction of the simulated personnel at the target moment and the preset normal moving direction of the robot is determined.

[0073] First, let the area occupied by all simulated personnel at the target time be q, and the total simulation area be s. Then the expression of the simulated personnel density at the tth time can be:

[0074] ; In the formula, represents the simulated personnel density at the tth moment, represents the area occupied by all simulated personnel at the tth moment, Represents the total simulation area at the tth moment.

[0075] Secondly, determine the distance between every two simulated personnel at the target time. Two simulated personnel are a pair. Perform negative correlation processing on the distance of each pair of simulated personnel. For example, use exp (-) to achieve negative correlation processing, and accumulate the distance value after each negative correlation processing to obtain the result value after accumulation processing, which is recorded as the proximity degree of the simulated personnel. The expression of the result value of the accumulation processing at the tth moment can be:

[0076] ; In the formula, represents the proximity of the simulated personnel at the tth moment, M represents the logarithm of the simulated personnel at the tth moment, and exp represents the exponential function with a natural constant as the base. It represents the distance between the mth pair of simulated personnel at the tth moment. The distance can be the position distance between the two simulated personnel, such as the Euclidean distance.

[0077] Then, the preset normal moving direction of the robot is set to 0 degrees. For any simulated person at the target time, the angle between the moving direction of the simulated person and the preset normal moving direction of the robot is determined, and the angle size of the angle is uniformly determined in the clockwise direction, which is recorded as the moving deviation angle. Among them, the moving deviation angle of the nth simulated person at the tth time can be recorded as .

[0078] In the second sub-step, the environmental complexity coefficient at the target time is determined by combining the simulated population density, the distribution fluctuation of the angle and the result value after cumulative processing; among them, the simulated population density, the distribution fluctuation of the angle and the result value are all positively correlated with the environmental complexity coefficient.

[0079] Exemplarily, the calculation formula of the environmental complexity coefficient at the t-th moment may be:

[0080] ; In the formula, represents the environmental complexity coefficient at the tth moment, represents the simulated personnel density at the tth moment, It represents the proximity of the simulated personnel at the tth moment, that is, the result value after the negative correlation value of the distance is accumulated. Var represents the variance function. represents the moving deviation angle of the nth simulated person at the tth moment, , N represents the number of simulated personnel at the tth moment.

[0081] In the calculation formula of the environmental complexity coefficient, the unit effect is not considered when the various calculation parts are multiplied, and only the corresponding numerical values ​​are taken to calculate the environmental complexity coefficient; The greater the density of simulated personnel and the more complex the environment, the more challenging the speed and path selection of the robot model in the simulation environment at the corresponding moment. At this time, the greater the communication pressure and the corresponding data feedback pressure, the greater the reliance on the high performance of the robot model. The greater the proximity of the simulated personnel, the closer the distance between each two simulated personnel, the higher the difficulty of path planning for the robot model, and the greater the environmental complexity coefficient; The larger the variance of the simulated personnel's movement deviation angle, the more complex the simulated personnel's movement trajectory and the higher the complexity of the environment, which is more unfavorable for the path planning and speed adjustment of the robot model.

[0082] The second step is to determine the rationality index of each model parameter corresponding to the robot model to be simulated at the current moment according to each model parameter sequence and the environmental complexity coefficient sequence.

[0083] It should be noted that the higher the complexity of the environment, the higher the difficulty of speed and path planning corresponding to the robot model in the simulation environment, the higher the error rate of the robot model in judging the model parameters such as speed, and the longer the machine operation time of the error rate, the greater the pressure on real-time communication, and the inability to process effective information in time. Therefore, in order to reduce the difficulty of communication and ensure the effectiveness of simulation data communication, it is necessary to increase the rationality of the robot model's judgment of different environments, specifically to analyze the logic of the physical environment, that is, the logical relationship between the complexity of the environment and the robot model parameters, and analyze and feedback and adjust according to the obtained relationship, and transmit the feedback results in real time to improve the transmission effectiveness of communication data.

[0084] The simulation environment diagram is as follows: Figure 3 As shown, in Figure 3 In the , when the crowd density increases, that is, the simulated personnel density increases, the optimal path planning length increases, and the number of turns increases, that is, the path turning points increase.

[0085] The second step of step S2 above can be Figure 4 The steps shown achieve:

[0086] S21, curve fitting is performed on each model parameter sequence and environmental complexity coefficient sequence to obtain respective fitting curves, the fitting curve corresponding to the environmental complexity coefficient sequence is used as a reference curve, and the fitting curve corresponding to the model parameter sequence is used as a comparison curve.

[0087] In this embodiment, in order to analyze the changing trends of model parameters and environmental parameters in the same period, the least square method can be used to perform curve fitting processing on each sequence to obtain each fitting curve; at the same time, in order to distinguish fitting curves of different properties, the fitting curve corresponding to the environmental complexity coefficient sequence is recorded as the reference curve, and the fitting curve corresponding to the model parameter sequence is recorded as the comparison curve, so as to facilitate the subsequent analysis of the logical relationship between each comparison curve and the reference curve. Among them, the implementation process of the least square method is a prior art, which is not within the protection scope of the present invention and will not be elaborated here.

[0088] S22, determining extreme value points in the reference curve and each comparison curve, forming an extreme value interval from curve intervals corresponding to two adjacent extreme value points, and obtaining each extreme value interval of the reference curve and each comparison curve.

[0089] In this embodiment, when the complexity of the simulation environment changes, the relevant model parameters of the robot model in the simulation environment will also be optimized accordingly. Every time the environmental complexity coefficient changes, the robot model needs to change accordingly to cope with the current environmental conditions. Therefore, in order to accurately analyze the rationality of model parameter judgment each time the environmental complexity changes, it is necessary to divide the curve using extreme points and divide it into intervals corresponding to two adjacent extreme points.

[0090] Specifically, the extreme points in the reference curve and each comparison curve are first determined, and the extreme points include maximum points and minimum points; then, the curve is divided using the extreme points on the curve as segmentation points to obtain the extreme value intervals corresponding to every two adjacent extreme value points, and the reference curve and the comparison curve each have their corresponding multiple extreme value intervals, and the extreme value intervals are local curve segments on the curve. Among them, the method for determining the extreme points is a prior art, which is not within the protection scope of the present invention and will not be elaborated in detail here.

[0091] S23, determining the rationality index of each model parameter corresponding to the robot model to be simulated at the current moment according to the slope value at each moment in each extreme value interval of the reference curve and each comparison curve.

[0092] It should be noted that two types of environmental parameters are selected as examples for explanation. Specifically, when the environmental complexity coefficient of the simulation environment increases, the robot model needs to reduce the real-time speed to avoid collisions and ensure the safety of pedestrians. At this time, the environmental complexity coefficient is set to be inversely proportional to the real-time speed; when the environmental complexity coefficient of the simulation environment increases, the robot model needs to use the path planning algorithm to find a reasonable and optimal path that avoids high density in a more complex environmental map. In order to ensure the safe distance between humans and machines, the environmental complexity coefficient is set to be proportional to the total length of the path planning. Among them, the real-time speed curve of the machine model and the change trend of the environmental complexity coefficient curve are inversely proportional.

[0093] The above step S23 can be performed by Figure 5 The steps shown achieve:

[0094] S231, determining a parameter rationality determination factor at each moment in each extreme value interval of each comparison curve according to the slope value at each moment in each extreme value interval of each comparison curve.

[0095] To facilitate understanding of this embodiment, any comparison curve is used as a target comparison curve, any extreme value interval with the same sequence number in the reference curve and the target comparison curve is used as the target extreme value interval, and the reasonable parameter determination factor at each moment in the target extreme value interval of the target comparison curve is determined. The specific implementation steps may include:

[0096] The product of the slope of the benchmark curve at each moment in the target extreme value interval and the slope of the target comparison curve at the corresponding moment in the target extreme value interval is calculated, and the slope product is processed using the sign function, and the result value after data processing is used as the reasonable determination factor of the parameter at each moment.

[0097] As an example, the calculation formula of the parameter reasonable determination factor at the vth moment in the uth extreme value interval of the i-th comparison curve can be:

[0098] ; In the formula, It represents the reasonable determination factor of the parameter at the vth moment in the uth extreme value interval of the ith comparison curve, sign represents the data sign function, It represents the slope of the i-th contrast curve at the v-th moment in the u-th extreme interval. Represents the slope of the reference curve at the vth moment in the uth extreme value interval.

[0099] In the calculation formula of the parameter rationality determination factor, the parameter rationality determination factor may take -1, 1 or 0. The parameter rationality determination factor can be used to determine whether the model parameters at the vth moment are adapted to the complexity of the simulation environment at the corresponding moment; for example, for the real-time speed of the robot model, the real-time speed is inversely proportional to the environmental complexity coefficient, so when the parameter rationality determination factor at the vth moment is -1, it means that the real-time speed setting of the robot model at the vth moment is reasonable, and when it is 1 or 0, it means that the real-time speed setting of the robot model at the vth moment is unreasonable.

[0100] It is worth noting that there is a situation where the number of data points in the u-th extreme value interval of the comparison curve is inconsistent with that in the u-th extreme value interval of the reference curve. In this case, the u-th extreme value interval with the smallest number of data points shall be taken as the standard. For example, there are 10 data points in the u-th extreme value interval of the comparison curve, and there are 8 data points in the u-th extreme value interval of the reference curve. In this case, the extreme value interval with 8 data points shall be taken as the standard, and 8 data points at the same time shall be determined in the u-th extreme value interval of the comparison curve for analysis.

[0101] S232, determining a parameter reasonable determination value for each extreme value interval of each comparison curve according to the parameter reasonable determination factor at each moment in each extreme value interval of each comparison curve and the number of moments in each extreme value interval.

[0102] After determining the reasonable determination factors of parameters at each moment, this embodiment needs to further quantify the reasonableness of the model parameters corresponding to the single change of the environmental complexity coefficient in combination with the reasonable determination factors of parameters at each moment. Specifically, if the reasonable determination factors of parameters at most moments in the extreme value interval satisfy the proportional relationship between the model parameters and the environmental complexity coefficient, it means that the model parameter settings of the extreme value interval are highly reasonable.

[0103] The above step S232 can be performed by Figure 6 The steps shown achieve:

[0104] S2321, based on the proportional relationship between the model parameters corresponding to the target comparison curve and the environmental complexity coefficient, a reasonable interval parameter determination model is constructed.

[0105] In this embodiment, the proportional relationship may be a direct proportional relationship or an inverse proportional relationship, and different proportional relationships may correspond to different expressions of the interval parameter reasonable determination model.

[0106] When the model parameters corresponding to the target comparison curve are proportional to the environmental complexity coefficient, the expression of the interval parameter reasonable judgment model is: ; In the formula, represents the reasonable judgment value of the parameter in the uth extreme value interval, exp represents the exponential function with the natural constant as the base, represents the number of moments in the u-th extreme value interval, v represents the sequence number of moments in the u-th extreme value interval, It represents the reasonable determination factor of the parameter at the vth moment in the uth extreme value interval.

[0107] In the expression of the interval parameter reasonable judgment model, when the model parameter corresponding to the target comparison curve is proportional to the environmental complexity coefficient, the more the number of moments when the parameter reasonable judgment factor is 1 in the u-th extreme value interval, the higher the parameter reasonable judgment value of the u-th extreme value interval, that is, the cumulative value of the parameter reasonable judgment factor in the u-th extreme value interval is more consistent with the number of moments in the u-th extreme value interval, that is, The smaller the value, the more reasonable the parameter is. The larger the value, the smaller the value. exp(-) is used to achieve negative correlation processing of data, and the value range of the processed data is between 0 and 1.

[0108] When the model parameters corresponding to the target comparison curve are inversely proportional to the environmental complexity coefficient, the expression of the interval parameter reasonable determination model is: .

[0109] In the expression of the interval parameter reasonable judgment model, when the model parameter corresponding to the target comparison curve is inversely proportional to the environmental complexity coefficient, the more the number of moments when the parameter reasonable judgment factor is -1 in the u-th extreme value interval, the higher the parameter reasonable judgment value of the u-th extreme value interval, that is, The smaller the value, the more reasonable the parameter is. The bigger.

[0110] S2322, the parameter reasonable determination factor and the number of moments at each moment in the target extreme value interval are input as input data into the interval parameter reasonable determination model to obtain the parameter reasonable determination value of the target extreme value interval of the target comparison curve.

[0111] In this embodiment, the parameter reasonable judgment factor and the number of moments at each moment in the target extreme value interval are taken as independent variables and substituted into the expression of the corresponding interval parameter reasonable judgment model. The parameter reasonable judgment value of the target extreme value interval can be obtained by calculation. The larger the parameter reasonable judgment value, the higher the reasonableness of the setting of the corresponding model parameters of the corresponding time period of the target extreme value interval.

[0112] S233, determining the rationality index of the model parameters corresponding to each comparison curve according to the parameter rationality judgment value of each extreme value interval of each comparison curve.

[0113] By referring to the calculation process of the reasonable judgment value of the parameter of the target extreme value interval of the target comparison curve, the reasonable judgment value of the parameter of each extreme value interval of each comparison curve can be obtained, which will not be repeated here. After obtaining the reasonable judgment value of the parameter of each extreme value interval of each comparison curve, this embodiment combines the reasonable judgment values ​​of the parameters of all extreme value intervals in the same comparison curve to determine the rationality index of the model parameters corresponding to the comparison curve, that is, to determine the rationality index of each model parameter of the robot model located in the simulation environment at the current moment. The rationality index can indicate the rationality of the setting of each model parameter of the robot model located in the simulation environment. The stronger the ability of the robot model to cope with complex simulation environments, the higher the rationality of the model parameters it sets. Taking the target comparison curve as an example to determine the rationality index of the model parameters corresponding to the target comparison curve, the implementation steps may include:

[0114] Specifically, the reasonable judgment value of the parameter of each extreme value interval of the target comparison curve is obtained, the cumulative value of the reasonable judgment value of the parameter of all extreme value intervals of the target comparison curve is calculated, and then the difference between the number of extreme value intervals of the target comparison curve and the cumulative value of the reasonable judgment value of the parameter is calculated to determine the rationality index of the model parameter corresponding to the target comparison curve; wherein, the difference is negatively correlated with the rationality index.

[0115] As an example, the calculation formula of the rationality index of the model parameters corresponding to the target comparison curve can be:

[0116] ; In the formula, It represents the rationality index of the model parameters corresponding to the target comparison curve, u represents the extreme value interval number of the target comparison curve, and U represents the number of extreme value intervals of the target comparison curve. represents the reasonable judgment value of the parameter of the uth extreme value interval of the target comparison curve, norm represents the normalization function, Represents a hyperparameter. The empirical value of the hyperparameter can be 0.01, which can be used to avoid the situation where the denominator of the fraction is zero. The implementer can set the hyperparameter size according to the actual situation.

[0117] In the calculation formula of rationality index, The value range is between 0 and 1. Less than or equal to U, when When it is infinitely close to U, it means that the reasonable judgment values ​​of the parameters in each extreme value interval of the target comparison curve are relatively large, and the higher the reasonableness of the setting of the model parameters corresponding to the target comparison curve, the higher the rationality index; the higher the rationality index, the more reasonable the relationship between the model parameters of the robot model and the environmental complexity of the simulation environment, that is, the changing relationship between the real-time speed of the robot model and the environmental complexity coefficient is satisfied; referring to the determination process of the rationality index of the model parameters corresponding to the target comparison curve, the rationality index of the model parameters corresponding to each comparison curve can be determined, and the obtained rationality index is used as the rationality index of each model parameter corresponding to the robot model to be simulated at the current moment.

[0118] It should be noted that each adjustment of the model parameters is implemented on the basis of the previous adjustment. When analyzing each model parameter corresponding to the robot model to be simulated at the current moment, the environmental parameters at the current moment have the most direct impact on the model, and the environmental parameters at other historical moments are only involved in the construction to facilitate the determination of the correlation characteristics between the environmental parameters and the model parameters, which improves the calculation accuracy of the rationality index of each model parameter at the current moment.

[0119] So far, this embodiment has obtained the rationality index of each model parameter corresponding to the robot model to be simulated at the current moment.

[0120] S3, according to the rationality index, the interactive transmission of adjustment information between the simulation end and the model end is carried out.

[0121] First of all, it should be noted that compared with randomly interacting all data in real time, a part of the model parameters with reasonable settings can be screened out through the rationality index of each model parameter at the current moment, and the remaining model parameters can be interacted in real time in the order of the rationality index. This can reduce the amount of data transmission to a certain extent and give priority to transmitting data with high effectiveness, so as to overcome the defect that a large amount of computing resources occupy the bandwidth in the process of real-time interaction between the robot model end and the simulation end, which is not conducive to the real-time interaction of important simulation information, and improve the effectiveness of data communication in the simulation system.

[0122] The above step S3 can be Figure 7The steps shown achieve:

[0123] S31, comparing the rationality index of each model parameter with a preset index threshold, and selecting the model parameter whose rationality index is less than the preset index threshold as the model parameter to be transmitted at the simulation end.

[0124] In this embodiment, the preset index threshold can be set to 0.9. The implementer can set the threshold value according to the actual situation without specific limitation. When the rationality index of a model parameter is not less than the preset index threshold, it can be determined that the model parameter is reasonable, and the necessity of interaction between the two ends of the model parameter is low, that is, the simulation end and the model end do not need to perform real-time data transmission;

[0125] On the contrary, when the rationality index of a model parameter is less than the preset index threshold, the model parameter can be judged as unreasonable and the model parameter is used as the model parameter to be transmitted at the simulation end, that is, feedback tuning is required through real-time communication. Among them, real-time communication is reflected in distributed transmission, that is, the transmission priority of the model parameters to be transmitted can be determined by the distributed transmission algorithm.

[0126] S32, using a distributed transmission algorithm, the model parameters to be transmitted of the simulation end are sent to the model end, and the model end determines the adjustment result of each model parameter to be transmitted through the received model parameters to be transmitted, and feeds back the adjustment result to the simulation end.

[0127] In this embodiment, the transmission order of the model parameters to be transmitted is determined according to the minimum round-trip delay priority algorithm, and the data packets of the model parameters to be transmitted are sequentially filled into the sending window of the sub-flow according to the transmission order to realize distributed transmission. After the model end receives the model parameters to be transmitted, it adjusts the model parameters to be transmitted based on the specific actual situation, and transmits the adjustment results to the simulation end in real time, so as to play the role of real-time transmission and communication interaction of the adjustment information between the simulation end and the model end.

[0128] Among them, the adjustment of the model parameters to be transmitted can be: reducing the real-time speed of the robot model, which can be set to reduce the original speed by 5% each time. The implementer can set the speed reduction size according to the specific actual situation; planning in the simulation environment through the robot model algorithm RRT (Rapidly-exploring Random Tree) path planning technology, and transmitting the planning results to the simulation end in real time to facilitate the adjustment of the robot model in the simulation environment. The implementation process of RRT path planning technology is a prior art and is not within the scope of protection of the present invention, and will not be elaborated here.

[0129] So far, this embodiment has completed the real-time and effective information interaction between the robot model end and the simulation environment simulation end at the current moment.

[0130] S4, iteratively adjust each environmental parameter, and repeat steps S1 to S3 until the simulation end condition is met.

[0131] In this embodiment, the machine model parameters are tuned by iteratively adjusting each environmental parameter to analyze its performance. Iterative environmental parameters refer to real-time updating of environmental parameters such as simulated personnel density at different times, and feedback of the corresponding model is performed based on the update results. The implementation process is the same as the above steps S1 to S3 until the simulation end condition is met. By changing the environmental parameters several times, the setting accuracy of the machine model parameters is increased, and its robustness is also increased.

[0132] The conditions for the end of simulation vary depending on the specific application, but are usually pre-set simulation time, event completion, achievement of specific conditions, or user manual operation triggering. These conditions ensure the control of the simulation process and the expected end behavior.

[0133] The present invention further provides a data communication platform for a simulation system, comprising a processor and a memory, wherein the processor is used for processing instructions stored in the memory to implement a data communication method for the simulation system.

[0134] For the data communication platform of the simulation system, the real-time process of communication feedback can be realized through the simulation system platform. The communication objects are the simulation end and the machine model end. The communication content is the optimized data, that is, the model parameters before and after adjustment. The specific development can include:

[0135] First of all, the communication system architecture includes a simulation end and a model end. The simulation end is responsible for simulating the crowd density and other relevant environmental parameters in the actual environment, and providing real-time environmental data simulation, that is, simulating the behavior of the robot in the simulation environment, including movement speed and path selection; the model end receives data from the simulation end and updates its model parameters (such as movement speed and path planning) in real time. Specifically, it adjusts its decision-making algorithm based on feedback, optimizes performance, and uses communication feedback to provide real-time processes.

[0136] Secondly, for data collection and transmission, the simulation end collects different types of environmental parameters in the simulation environment in real time, and sends the collected data to the model end through a stable communication protocol; for data reception and processing, the model end receives data and analyzes the data through the built-in data processing module, where data processing includes extracting environmental features, analyzing crowd behavior patterns, and evaluating the effectiveness of the current robot model's behavior strategy.

[0137] Then, for feedback and parameter adjustment, based on the results of data processing, the model side generates adjustment suggestions, such as adjusting speed settings or modifying path planning, and these adjustment suggestions are then fed back to the simulation side through the communication system.

[0138] Finally, for the response and adjustment simulation of the simulation end, after receiving the adjustment suggestions, the simulation end updates the model parameters of the robot model in the simulation environment, and then the simulation end simulates the updated robot behavior and observes its performance in the modified environment to ensure that the adjustment is effective.

[0139] It should be noted that the purpose of establishing a communication platform includes: ensuring the real-time transmission of simulation data, so that all data and adjustment suggestions can be quickly and in real time transmitted and processed to reduce delays; ensuring the accuracy of data and the reliability of processing results to ensure effective optimization of robot performance; protecting all transmitted data to avoid potential security threats affecting simulation and actual operations. In short, through the data communication platform of the present invention, efficient interaction between the simulation platform and the machine model end can be achieved, further improving the adaptability and performance of the robot in crowded environments.

[0140] The present invention also provides a computer-readable storage medium, which stores computer program codes. When the computer program codes are run on a computer, the computer executes a data communication method for a simulation system.

[0141] This is the end of the present invention.

[0142] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A data communication method for a simulation system, characterized in that: The following steps are involved: S1: obtaining a plurality of environmental parameter sequences corresponding to the multi-mode physical simulation environment in the current period, and a plurality of model parameter sequences corresponding to the robot model to be simulated integrated in the multi-mode physical simulation environment; wherein the types of model parameters include the real-time speed of the robot model, the total length of the path planning, and the number of turning points; S2: According to the environmental parameters at each moment in each environmental parameter sequence, determine the environmental complexity coefficient at each moment in the current period to form an environmental complexity coefficient sequence; according to each model parameter sequence and the environmental complexity coefficient sequence, determine the rationality index of each model parameter corresponding to the robot model to be simulated at the current moment; S3: performing interactive transmission of adjustment information between the simulation end and the model end according to the rationality index; S4: Iteratively adjust each environmental parameter and repeat steps S1 to S3 until the simulation end condition is met; The step of determining the environmental complexity coefficient at each moment in the current period according to the environmental parameters at each moment in each environmental parameter sequence includes: The environmental parameters include the density of simulated personnel, the distance between any two simulated personnel, and the moving direction of each simulated personnel; taking any moment as the target moment, and determining the environmental complexity coefficient at the target moment according to the density of simulated personnel, the distance between any two simulated personnel, and the moving direction of each simulated personnel at the target moment; Determining the environmental complexity coefficient at the target time according to the simulated personnel density at the target time, the distance between any two simulated personnel, and the moving direction of each simulated personnel includes: The size of the total simulation area occupied by all simulated personnel at the target time is taken as the simulated personnel density; all negative correlation values ​​of the distances are accumulated to obtain the result value after the accumulation processing; and the angle between the moving direction of the simulated personnel at the target time and the preset normal moving direction of the robot is determined; Determine the environmental complexity coefficient at the target time by combining the simulated personnel density at the target time, the distribution fluctuation of the angle, and the result value after the accumulation processing; wherein the simulated personnel density, the distribution fluctuation of the angle, and the result value are all positively correlated with the environmental complexity coefficient; The method of determining the rationality index of each model parameter corresponding to the robot model to be simulated at the current moment according to each model parameter sequence and the environmental complexity coefficient sequence includes: Each model parameter sequence and environmental complexity coefficient sequence is subjected to curve fitting to obtain various fitting curves, the fitting curve corresponding to the environmental complexity coefficient sequence is used as the reference curve, and the fitting curve corresponding to the model parameter sequence is used as the comparison curve; Determine the extreme value points in the reference curve and each comparison curve, form an extreme value interval with the curve intervals corresponding to two adjacent extreme value points, and obtain each extreme value interval of the reference curve and each comparison curve; According to the slope value at each moment in each extreme value interval of the reference curve and each comparison curve, the rationality index of each model parameter corresponding to the robot model to be simulated at the current moment is determined; The method of determining the rationality index of each model parameter corresponding to the robot model to be simulated at the current moment according to the slope value at each moment in each extreme value interval of the reference curve and each comparison curve comprises: Take any comparison curve as the target comparison curve, and take any extreme value interval with the same sequence number in the reference curve and the target comparison curve as the target extreme value interval; calculate the product of the slope at each moment in the target extreme value interval of the reference curve and the slope at the corresponding moment in the target extreme value interval of the target comparison curve, and use the sign function to process the slope product, and use the result value after data processing as the reasonable parameter determination factor at each moment; According to the parameter reasonable determination factor and the number of moments at each moment in the target extreme value interval, the parameter reasonable determination value of the target extreme value interval of the target comparison curve is determined; Obtaining the reasonable judgment value of the parameter of each extreme value interval, calculating the cumulative value of the reasonable judgment value of the parameter of all extreme value intervals of the target comparison curve, and then calculating the difference between the number of extreme value intervals of the target comparison curve and the cumulative value of the reasonable judgment value of the parameter, and determining the rationality index of the model parameter corresponding to the target comparison curve; wherein the difference is negatively correlated with the rationality index; The method of determining the reasonable determination value of the parameter of the target extreme value interval of the target comparison curve according to the reasonable determination factor of the parameter at each moment in the target extreme value interval and the number of moments includes: Combined with the proportional relationship between the model parameters corresponding to the target comparison curve and the environmental complexity coefficient, a reasonable interval parameter determination model is constructed; the reasonable parameter determination factor and the number of moments at each moment in the target extreme value interval are used as input data and input into the reasonable interval parameter determination model to obtain the reasonable parameter determination value of the target extreme value interval of the target comparison curve; The steps of obtaining the reasonable interval parameter determination model include: When the model parameters corresponding to the target comparison curve are proportional to the environmental complexity coefficient, the expression of the interval parameter reasonable judgment model is: ; In the formula, represents the reasonable judgment value of the parameter in the uth extreme value interval, exp represents the exponential function with the natural constant as the base, represents the number of moments in the u-th extreme value interval, v represents the sequence number of moments in the u-th extreme value interval, Represents the reasonable determination factor of the parameter at the vth moment in the uth extreme value interval; When the model parameters corresponding to the target comparison curve are inversely proportional to the environmental complexity coefficient, the expression of the interval parameter reasonable determination model is: .

2. The data communication method of a simulation system according to claim 1, characterized in that: The interactive transmission of adjustment information between the simulation end and the model end according to the rationality index includes: Compare the rationality index of each model parameter with a preset index threshold, and select the model parameter whose rationality index is less than the preset index threshold as the model parameter to be transmitted at the simulation end; The model parameters to be transmitted at the simulation end are sent to the model end by using a distributed transmission algorithm. The model end determines the adjustment result of each model parameter to be transmitted by using the received model parameters to be transmitted, and feeds back the adjustment result to the simulation end.

3. A data communication platform for a simulation system, characterized in that: It comprises a processor and a memory, wherein the processor is used to process instructions stored in the memory to implement a data communication method for a simulation system as claimed in any one of claims 1 to 2.

4. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program codes, and when the computer program codes are executed on a computer, the computer is enabled to execute a data communication method for a simulation system as claimed in any one of claims 1 to 2.

Images