Intelligent distribution double-screen interaction method and terminal

Abstract

The invention relates to the technical field of video terminal interaction, in particular to an intelligent distribution double-screen interaction method and a terminal. Comprising the following steps: S1, acquiring vehicle data information on an operation screen, and generating a historical data information sample set; weight division is carried out on feature vector importance in a vehicle driving state map, a weighted total score of each evaluation index is calculated, the weighted total score is compared with a preset evaluation threshold value, whether an avoidance mechanism is started or not is judged according to a comparison result, corresponding driving route information is extracted, and the vehicle driving state map is determined. The driving route information is sent to the display screen of the vehicle through the data transmission module, a driver can see the recommended driving route on the display screen, observation is facilitated, the driving strategy of the driver can be adjusted in time according to the planned route, the vehicle can smoothly pass through the obstacle, and the accuracy of the vehicle passing through the obstacle is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of video terminal interaction, and in particular to an intelligently allocated dual-screen interaction method and terminal. Background Art

[0002] Coal mine double-headed vehicles usually refer to double-headed mining vehicles or double-headed engineering vehicles. They play a vital role in the field of mining, especially coal mining. In order to facilitate the driver to control the vehicle, cameras and ultrasonic radars are installed around the vehicle to display key information during the vehicle's driving process in real time on two display screens, thereby realizing effective transmission and monitoring of information.

[0003] When a coal mine double-headed vehicle is operating in a mine, the prior art only relies on sensor feedback data to analyze the distance between the vehicle and obstacles, and then plans the vehicle's route. This is single-minded, and traditional vehicle navigation systems often rely on fixed map data and path planning. However, in complex road environments, for example, due to the complex internal environment of the mine, smoke and dust will be generated during the process of the vehicle dumping coal. The smoke and dust will cover the camera and form an obstruction, causing the operating screen inside the vehicle to be blurred, and the picture on the interactive display screen is also unclear. At this time, the driver can only rely on traditional driving experience, sensors, and route markings in the mine to control the vehicle. That is, these navigation systems cannot flexibly respond to obstacles on the route, resulting in the accuracy and safety of the vehicle passing through obstacles. Summary of the invention

[0004] The purpose of the present invention is to provide a dual-screen interaction method and terminal with intelligent allocation to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides a dual-screen interaction method with intelligent allocation, comprising the following method steps:

[0006] S1. Obtain vehicle data information on the operation screen and generate a sample set of historical data information;

[0007] S2. Based on the ultrasonic radar, the vehicle position status and obstacle information are obtained. The phase difference angle measurement algorithm is used to calculate the signal path difference, phase difference and the relative angle between the vehicle and the obstacle according to the distance parameters between the ultrasonic radars. The vehicle driving status map is constructed in combination with the obtained vehicle data information.

[0008] S3, storing the vehicle driving state map in the vehicle driving state material library, and establishing an indexing mechanism, and then combining the dynamic time warping algorithm and the vehicle driving state similarity algorithm, comparing the vehicle driving state map with the historical vehicle driving state map, predicting the future driving state of the vehicle based on the comparison result, and weighting the importance of the feature vectors in the vehicle driving state map, calculating the weighted total score of the evaluation indicators, comparing the weighted total score with the preset evaluation threshold, and judging whether to start the avoidance mechanism according to the comparison result;

[0009] S4. Based on the comparison results, the weighted total score in the historical vehicle driving status map is obtained, and the weighted total score is screened. The historical planned route corresponding to the weighted total score after screening is executed. Based on the feedback in avoiding obstacles, the dynamic time warping algorithm and the relative angle similarity algorithm are optimized, and the stored historical vehicle driving status data is updated in combination with the optimization results.

[0010] As a further improvement of the technical solution, based on the phase difference angle measurement algorithm and according to the distance parameter between the ultrasonic radars, the signal path difference, phase difference and the relative angle between the vehicle and the obstacle are calculated, as shown below:

[0011] The relationship between the phase difference ΔX and the path difference ΔR, the ultrasonic wavelength λ and the antenna spacing d can be expressed as

[0012]

[0013] ΔR=d×sinθ, θ is the angle between the antenna and the obstacle, so the phase difference ΔX can be expressed as

[0014]

[0015] The path difference ΔR is the difference in distance between the two antennas.

[0016] As a further improvement of this technical solution, the steps of constructing a vehicle driving state map are as follows:

[0017] The relative angle θ, vehicle speed V, distance D between the vehicle and the obstacle, and wheel steering speed V1 are used as two-dimensional data of the map, and the coordinate axis is determined. The relative angle θ, vehicle speed V, distance D between the vehicle and the obstacle, and wheel steering speed V1 are used as the Y axis of the map, and the time point node is used as the X axis to construct a two-dimensional vehicle driving state map, and the collected data points are marked on the map in real time;

[0018] Among them, the relative angle θ, vehicle speed V, distance D between the vehicle and the obstacle, and wheel steering speed V1 are connected according to the time node to form a trajectory curve

[0019] As a further improvement of this technical solution, the method of constructing the index mechanism is as follows:

[0020] Using an indexing method based on a hash algorithm, the feature vector of the vehicle driving state map is hashed to obtain the corresponding hash value;

[0021] A hash index table is constructed by using the hash value as the index key and the storage address of the graph in the vehicle driving status material library as the index value.

[0022] As a further improvement of the technical solution, a dynamic time warping algorithm is used to warp the time points of the vehicle driving state graph and the historical vehicle driving state graph in the data information sample set, and find the optimal matching path of the two graphs on the time axis;

[0023] Along the optimal matching path, the similarity of the vehicle driving state at the corresponding time point is calculated. For the calculation of the similarity of the vehicle driving state, the similarity calculation formula based on the Euclidean distance is adopted to calculate the distribution similarity of the relative angle θ, the vehicle speed V, the distance D between the vehicle and the obstacle, and the wheel steering speed V1 respectively.

[0024] As a further improvement of the technical solution, similarity range values ​​are set respectively, and weights are allocated based on the importance of the similarity range value and the relative angle θ, the vehicle speed V, the distance D between the vehicle and the obstacle, and the wheel steering speed V1. A weighted total score is calculated based on the weight of each evaluation index and the performance of the vehicle on the index. The weighted total score can be calculated by the following formula:

[0025]

[0026] Among them, Q is the weighted total score, n is the number of evaluation indicators, and f i is the score of indicator i, g i The weight of indicator i is used to obtain f i When calculating the score of a feature vector, the corresponding preset threshold interval can be set by the feature vector, and the score proportion of the feature vector falling into different preset threshold intervals can be distributed.

[0027] As a further improvement of the technical solution, the importance is divided as follows: relative angle θ> vehicle speed V> distance D between vehicle and obstacle> wheel steering speed V1;

[0028] W s =W V +W d +W θ +W w ;

[0029] W V , W d , Wθ , W w The weights of the relative angle, vehicle speed, distance between the vehicle and the obstacle, and wheel steering speed are respectively corresponding, and the sum of the weights is 1. The weighted total score Q is compared with the set comprehensive evaluation threshold. If the weighted total score is higher than the threshold, it is considered that the vehicle has the conditions to pass the obstacle and the vehicle is driving normally; if it is lower than the threshold, it is considered that the vehicle does not have the conditions to pass the obstacle.

[0030] As a further improvement of the technical solution, when processing the image input from the operation screen to the display screen, the point operation algorithm in the image enhancement technology is used to enhance the image in the image sample, as follows:

[0031] Change the contrast of the image by linearly stretching or compressing the grayscale value of the image;

[0032] h(j│k)=a×q(j│k)+b;

[0033] Among them, q(j│k) is the grayscale value of the original image, h(j│k) is the grayscale value after transformation, and a and b are the coefficients of linear transformation.

[0034] The second object of the invention is to provide a dual-screen interactive terminal for operating the intelligent allocation as described above, including the interactive method as described above, in which the operating screen, the display screen and the vehicle system establish a communication connection, and a right interactive module is set between the vehicle system and the display screen. The interactive module is used to respond to the user's interactive instructions on the operating screen and map the interactive operations generated on the operating screen to the display screen.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] In the dual-screen interactive method of intelligent allocation, by weighting the importance of feature vectors in the vehicle driving state map, the weighted total score of the evaluation indicators is calculated, the weighted total score is compared with the preset evaluation threshold, and whether to start the avoidance mechanism is determined according to the comparison result. When the system determines that the vehicle does not have the conditions to pass through the obstacle, the weighted total score of the current driving state is obtained, and the records with the weighted total score greater than the set threshold are screened out from the historical vehicle driving state map set. Among these records, the record closest to the weighted total score of the current driving state is found, and the corresponding driving route information is extracted from the closest record, and the driving route information is sent to the vehicle display screen through the data transmission module. The driver can see the recommended driving route on the display screen for easy observation, and can adjust his driving strategy in time according to the planned route, so that the vehicle can pass through the obstacle smoothly, thereby improving the accuracy of the vehicle passing the obstacle. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 The figure is a flow chart of the intelligently allocated dual-screen interaction method of the present invention. DETAILED DESCRIPTION

[0038] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] like Figure 1 As shown, a dual-screen interaction method with intelligent allocation is provided, including the following method steps:

[0040] S1. Obtain vehicle data information on the operation screen and generate a sample set of historical data information;

[0041] S2. Based on the ultrasonic radar, the vehicle position status and obstacle information are obtained. The phase difference angle measurement algorithm is used to calculate the signal path difference, phase difference and the relative angle between the vehicle and the obstacle according to the distance parameters between the ultrasonic radars. The vehicle driving status map is constructed in combination with the obtained vehicle data information.

[0042] S3, storing the vehicle driving state map in the vehicle driving state material library, and establishing an indexing mechanism, and then combining the dynamic time warping algorithm and the vehicle driving state similarity algorithm, comparing the vehicle driving state map with the historical vehicle driving state map, predicting the future driving state of the vehicle based on the comparison result, and weighting the importance of the feature vectors in the vehicle driving state map, calculating the weighted total score of the evaluation indicators, comparing the weighted total score with the preset evaluation threshold, and judging whether to start the avoidance mechanism according to the comparison result;

[0043] S4. Based on the comparison results, the weighted total score in the historical vehicle driving status map is obtained, and the weighted total score is screened. The historical planned route corresponding to the weighted total score after screening is executed. Based on the feedback in avoiding obstacles, the dynamic time warping algorithm and the relative angle similarity algorithm are optimized, and the stored historical vehicle driving status data is updated in combination with the optimization results.

[0044] Among them, the vehicle data information obtained in S1 includes vehicle parameters, radar position, vehicle speed, the distance between the vehicle and the obstacle, and the wheel steering speed. The vehicle speed V can be obtained through the vehicle speed sensor, and as for the distance D between the vehicle and the obstacle, the radar transmits radio waves and receives the reflected signals, and the distance is determined by calculating the time difference of the signals.

[0045] According to the acquired vehicle data information, based on the phase difference angle measurement algorithm, and according to the distance parameters between the ultrasonic radars, the signal path difference, phase difference and the relative angle between the vehicle and the obstacle are calculated, as shown below:

[0046] The relationship between the phase difference ΔX and the path difference ΔR, the ultrasonic wavelength λ and the antenna spacing d can be expressed as

[0047]

[0048] ΔR=d×sinθ, θ is the angle between the antenna and the obstacle, so the phase difference ΔX can be expressed as

[0049]

[0050] The path difference ΔR is the difference in distance between the two antennas.

[0051] Among them, the phase difference method angle measurement uses the phase difference between ultrasonic echo signals received by multiple antennas (or receivers) to measure the angle. For example, there are two antennas A1 and B1, the distance between them is d, and both receive ultrasonic echo signals from the same obstacle. Due to the distance d between antennas A1 and B1, the signals they receive will have a path difference ΔR, resulting in a phase difference ΔX.

[0052] Assume that the distance between antennas A1 and B1 is d = 1 meter, the speed of ultrasound in air is v = 340 m / s, and the frequency of the signal is f = 40 kHz, so the wavelength is Meters, if the distance d from the obstacle to antenna A1 A1 = 2 meters, distance d to antenna B1 B1 =2.5 meters, then the path difference ΔR=2-2.5=-0.5 meters (the negative sign indicates that the signal received by A1 arrives earlier than that received by B1).

[0053] Therefore, the phase difference

[0054] It should be noted that the phase difference is usually expressed in degrees or radians. Here, it is converted to degrees and rounded to one decimal place for ease of understanding. Since the phase difference is periodic, -396.6° can also be expressed as -396.6+360=-9.6 degrees, but in this example, we keep the negative sign to indicate that the signal received by A1 arrives earlier than B1.

[0055]

[0056] Similarly, the angle θ can also be positive or negative, depending on the sign of the path difference ΔR. In this example, -10° means that the obstacle is located on the extension line of the line connecting antennas A1 and B1, and is about 10 degrees to the side of A1.

[0057] The steps to construct a vehicle driving state map are as follows:

[0058] The relative angle θ, vehicle speed V, distance D between the vehicle and the obstacle, and wheel steering speed V1 are used as two-dimensional data of the map, and the coordinate axis is determined. The relative angle θ, vehicle speed V, distance D between the vehicle and the obstacle, and wheel steering speed V1 are used as the Y axis of the map, and the time point node is used as the X axis to construct a two-dimensional vehicle driving state map, and the collected data points are marked on the map in real time;

[0059] Among them, the relative angle θ, the vehicle speed V, the distance D between the vehicle and the obstacle, and the wheel steering speed V1 are connected respectively according to the time node to form a trajectory broken line.

[0060] For example, when a vehicle is driving, its driving state changes over time as follows:

[0061] At the initial moment, the vehicle speed is 0 km / h, the relative angle is 0 degrees (i.e. the vehicle is parallel to the obstacle), and the distance from the obstacle is 10 meters. Then, the vehicle starts to accelerate and turns right. During the acceleration process, the speed gradually increases, the relative angle shifts to the right, and the distance from the obstacle changes due to the turn. Finally, the vehicle reaches a certain stable speed and continues to drive while maintaining a certain relative angle and distance from the obstacle.

[0062] For example:

[0063] The X-axis is the time point: 0 seconds, 1 second, ..., 10 seconds;

[0064] The Y-axes correspond to:

[0065] Vehicle speed: the speed value at each time point, for example [50,55,60,...,100]km / h;

[0066] Relative angle: the angle value at each time point, for example [0,5,10,...,45] degrees;

[0067] Distance: the distance value at each time point, for example [10,8,6,...,2] meters;

[0068] Steering speed: the steering speed value at each time point, for example [0.5, 1.0, 1.5, ..., 3.0] degrees / second;

[0069] The vehicle speed, relative angle, distance, and steering speed at each time point are connected separately to intuitively understand the state changes of the vehicle during driving.

[0070] After the above-mentioned atlas is established, it is stored in the vehicle driving state material library, and an index mechanism is established to facilitate later search. The method of constructing the index mechanism is as follows:

[0071] Using an indexing method based on a hash algorithm, the feature vector of the vehicle driving state map is hashed to obtain the corresponding hash value;

[0072] A hash index table is constructed by using the hash value as the index key and the storage address of the graph in the vehicle driving status material library as the index value.

[0073] The specific steps are as follows: first, for each vehicle driving state map, extract its feature vector, then use a hash algorithm (such as MD5, SHA-256, etc.) to hash the feature vector to obtain a hash value of a fixed length, and use the hash value as the index key to create a hash index table, where the index key is the hash value, and the index value is the storage address of the vehicle in the driving state material library;

[0074] For example, there are three vehicle driving state graphs and their feature vectors:

[0075] Spectrum 1: eigenvector = [30°, 10, 10, 0.2];

[0076] Spectrum 2: eigenvector = [40°, 20, 20, 0.3];

[0077] Spectrum 3: eigenvector = [50°, 30, 30, 0.4];

[0078] Use the hash algorithm to calculate the feature vector, get the hash value, and build the hash index table as follows:

[0079] Hash value (index key) storage address (index value);

[0080] md5(30°,10,10,0.2)=a1b2c3d4e5f6...→ / path / to / vehicle_state_1.png;

[0081] md5(40°,20,20,0.3)=b2a1d3c4e5f6...→ / path / to / vehicle_state_2.png;

[0082] md5(50°,30,30,0.4)=c3b2a1d4e5f6...→ / path / to / vehicle_state_3.png;

[0083] When you need to retrieve a vehicle map under a specific driving state, you can perform the following steps:

[0084] 1. Extract the feature vector of the target driving state.

[0085] 2. Perform hash calculation on the feature vector to obtain the hash value.

[0086] 3. Find the storage address corresponding to the hash value in the hash index table.

[0087] 4. Obtain the corresponding vehicle map from the vehicle driving status material library according to the storage address.

[0088] Based on the comparison of historical data, a neural network prediction model can be established to predict the future driving status and possible problems of the vehicle. The system can issue early warning information to remind the driver or maintenance personnel to conduct timely inspection and maintenance. For example, if the performance of the vehicle often deteriorates within a certain period of time, maintenance or adjustment can be carried out in advance before that period of time.

[0089] Furthermore, a dynamic time warping algorithm is used to warp the time points of the vehicle driving state graph and the historical vehicle driving state graph in the data information sample set, and find the optimal matching path of the two graphs on the time axis;

[0090] Suppose we have two vehicle driving status graphs, represented as time series A and time series B respectively:

[0091] Time series C (current vehicle driving state map): C1, C2, C3, ..., Cn

[0092] Time series D (historical vehicle driving status map): D1, D2, D3, ..., Dm

[0093] Among them, each element Ci or Dj can represent the relative angle θ of the vehicle at a certain point in time, the vehicle speed V, the distance D between the vehicle and the obstacle, the wheel steering speed V1 and other information;

[0094] The DTW algorithm is used to find the optimal matching path between two time series so that the cumulative distance between them is minimized. This path represents how the time series of the current vehicle driving state map and each sample in the historical vehicle driving state map set correspond on the time axis to minimize the difference between them.

[0095] Along the optimal matching path, the similarity of the vehicle driving state at the corresponding time point is calculated. For the calculation of the similarity of the vehicle driving state, the similarity calculation formula based on the Euclidean distance is adopted to calculate the distribution similarity of the relative angle θ, the vehicle speed V, the distance D between the vehicle and the obstacle, and the wheel steering speed V1 respectively.

[0096] Assuming there are two states S1 and S2, each state contains the values ​​of the above four evaluation indicators, these values ​​can be regarded as points in a four-dimensional space, and the Euclidean distance formula is used to calculate the distance between the two points:

[0097]

[0098] Among them, V1-V2, d1-d2, θ1-θ2, and w1-w2 are the differences between the two states in vehicle speed, distance between the vehicle and the obstacle, relative angle, and wheel steering speed, respectively.

[0099] Furthermore, similarity range values ​​are set respectively, and weights are allocated based on the similarity range values ​​and the importance of the relative angle θ, the vehicle speed V, the distance D between the vehicle and the obstacle, and the wheel steering speed V1. A weighted total score is calculated based on the weight of each evaluation indicator and the performance of the vehicle on the indicator. The weighted total score can be calculated using the following formula:

[0100]

[0101] Among them, Q is the weighted total score, n is the number of evaluation indicators, and f i is the score of indicator i, g i The weight of indicator i is used to obtain f i When the score is obtained, the corresponding preset threshold interval can be set by the feature vector, and the score proportion of the feature vector falling into different preset threshold intervals can be distributed;

[0102] For example: v =30km / h, the corresponding preset threshold interval is [0-30]km / h, the weight allocation ratio is 70%, the preset threshold interval is [30-50]km / h, the weight allocation ratio is 20%, the preset threshold interval is [50-70]km / h, the weight allocation ratio is 10%, and the other feature vectors are the same as above and will not be repeated here.

[0103] Importance division: relative angle θ> vehicle speed V> distance D between vehicle and obstacle> wheel steering speed V1;

[0104] W s =W V +W d +W θ +W w;

[0105] W V , W d , W θ , W w The weights of the relative angle, vehicle speed, distance between the vehicle and the obstacle, and wheel steering speed are respectively corresponding, and the sum of the weights is 1. The weighted total score Q is compared with the set comprehensive evaluation threshold. If the weighted total score is higher than the threshold, it is considered that the vehicle has the conditions to pass the obstacle and the vehicle is driving normally; if it is lower than the threshold, it is considered that the vehicle does not have the conditions to pass the obstacle.

[0106] For example, with the following weight and threshold settings:

[0107] Weight: W V =0.3,W d =0.2,W θ =0.3,W w =0.2;

[0108] Comprehensive evaluation threshold: T = 0.7;

[0109] The calculated similarities are: the relative angle θ, the vehicle speed V, the distance D between the vehicle and the obstacle, and the wheel steering speed V1, the similarities correspond to 0.8, 0.6, 0.9, and 0.7;

[0110] Then Q = 0.3 × 0.8 + 0.2 × 0.6 + 0.3 × 0.9 + 0.2 × 0.7 = 0.77;

[0111] Since 0.77>0.7, it is judged that the vehicle is qualified to pass the obstacle and the vehicle drives normally. On the contrary, if Q is less than 0.7, it is judged that the vehicle is not qualified to pass the obstacle, and then the avoidance mechanism is executed to execute the planned driving route.

[0112] That is: when the weighted total score is lower than the comprehensive evaluation threshold, the weighted total score in the historical vehicle driving status map is obtained, and the records whose weighted total score is greater than the comprehensive evaluation threshold and closest to the weighted total score are filtered out from the weighted total score, and the driving route corresponding to the record is transmitted to the operation screen and display screen through data.

[0113] By weighting the importance of the feature vectors in the vehicle driving state map, the weighted total score of the evaluation indicators is calculated, and the weighted total score is compared with the preset evaluation threshold. According to the comparison result, it is determined whether to start the avoidance mechanism. When the system determines that the vehicle does not have the conditions to pass through the obstacle, the weighted total score of the current driving state is obtained, and the records with the weighted total score greater than the set threshold are screened out from the historical vehicle driving state map set. Among these records, the record closest to the weighted total score of the current driving state is found, and the corresponding driving route information is extracted from the closest record. The driving route information is sent to the vehicle display screen through the data transmission module. The driver can see the recommended driving route on the display screen for easy observation, and can adjust his driving strategy in time according to the planned route, so that the vehicle can pass through obstacles smoothly, thereby improving the accuracy of the vehicle passing obstacles.

[0114] After the driver follows the recommended route, the system collects feedback data from the vehicle during the actual driving process, including whether obstacles are successfully avoided, driving time, driving distance, feature vectors, etc. Based on the collected feedback data, the dynamic time warping algorithm (DTW) and the relative angle similarity algorithm are optimized. For example, the path search strategy in the DTW algorithm can be adjusted, or the calculation method of relative angle similarity can be improved to improve the accuracy of the algorithm in complex scenarios. The optimized algorithm is applied to new driving status data to generate a new weighted total score and driving route information, which is stored in the database as a new historical record for future comparison and reference.

[0115] Moreover, when the image on the operation screen is interactively transferred to the display screen, in order to ensure the clarity of the planned route on the display screen, first, when processing the image input from the operation screen to the display screen, the point operation algorithm in the image enhancement technology is used to enhance the image in the image sample, as follows:

[0116] Change the contrast of the image by linearly stretching or compressing the grayscale value of the image;

[0117] h(j│k)=a×q(j│k)+b;

[0118] Among them, q(j│k) is the grayscale value of the original image, h(j│k) is the grayscale value after transformation, and a and b are the coefficients of linear transformation.

[0119] For example, the grayscale value range of the original image is between 50 and 200, and the target grayscale value range is between 0 and 255. Then, you can choose b=-50×1.7=-85. In this way, the transformed grayscale value h(j│k)=1.7×q(j│k)-85 can stretch the grayscale value range of the original image to between 0 and 255, thereby enhancing the contrast of the image and improving the visual effect of the image.

[0120] By processing the acquired images, they can be clearly displayed on the display screen, allowing the operator to make timely adjustments to the vehicle status according to the planned route.

[0121] The second object of the present invention is to provide an intelligently allocated dual-screen interactive terminal, including the interactive method as described above, in which the operating screen, the display screen and the vehicle system establish a communication connection, and a right interactive module is set between the vehicle system and the display screen. The interactive module is used to respond to the user's interactive instructions on the operating screen and map the interactive operations generated on the operating screen to the display screen.

[0122] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. An intelligently allocated dual-screen interaction method, characterized in that: The method comprises the following steps: S1. Obtain vehicle data information on the operation screen and generate a sample set of historical data information; S2. Based on the ultrasonic radar, the vehicle position status and obstacle information are obtained. The phase difference angle measurement algorithm is used to calculate the signal path difference, phase difference and the relative angle between the vehicle and the obstacle according to the distance parameters between the ultrasonic radars. The vehicle driving status map is constructed in combination with the obtained vehicle data information. S3, storing the vehicle driving state map in the vehicle driving state material library, and establishing an indexing mechanism, and then combining the dynamic time warping algorithm and the vehicle driving state similarity algorithm, comparing the vehicle driving state map with the historical vehicle driving state map, predicting the future driving state of the vehicle based on the comparison result, and weighting the importance of the feature vectors in the vehicle driving state map, calculating the weighted total score of the evaluation indicators, comparing the weighted total score with the preset evaluation threshold, and judging whether to start the avoidance mechanism according to the comparison result; S4. Based on the comparison results, the weighted total score in the historical vehicle driving status map is obtained, and the weighted total score is screened. The historical planned route corresponding to the weighted total score after screening is executed. Based on the feedback in avoiding obstacles, the dynamic time warping algorithm and the relative angle similarity algorithm are optimized, and the stored historical vehicle driving status data is updated in combination with the optimization results.

2. The dual-screen interaction method of intelligent allocation according to claim 1, characterized in that: Based on the phase difference angle measurement algorithm, and according to the distance parameters between the ultrasonic radars, the signal path difference, phase difference and the relative angle between the vehicle and the obstacle are calculated, as shown below: The relationship between the phase difference ΔX and the path difference ΔR, the ultrasonic wavelength λ and the antenna spacing d can be expressed as ΔR=d×sinθ, θ is the angle between the antenna and the obstacle, so the phase difference ΔX can be expressed as The path difference ΔR is the difference in distance between the two antennas.

3. The dual-screen interaction method of intelligent allocation according to claim 2 is characterized in that: The steps to construct a vehicle driving state map are as follows: The relative angle θ, vehicle speed V, distance D between the vehicle and the obstacle, and wheel steering speed V1 are used as two-dimensional data of the map, and the coordinate axis is determined. The relative angle θ, vehicle speed V, distance D between the vehicle and the obstacle, and wheel steering speed V1 are used as the Y axis of the map, and the time point node is used as the X axis to construct a two-dimensional vehicle driving state map, and the collected data points are marked on the map in real time; Among them, the relative angle θ, the vehicle speed V, the distance D between the vehicle and the obstacle, and the wheel steering speed V1 are connected respectively according to the time node to form a trajectory broken line.

4. The dual-screen interaction method of intelligent allocation according to claim 3 is characterized in that: The method of constructing the index organization is as follows: Using an indexing method based on a hash algorithm, the feature vector of the vehicle driving state map is hashed to obtain the corresponding hash value; A hash index table is constructed by using the hash value as the index key and the storage address of the graph in the vehicle driving status material library as the index value.

5. The dual-screen interaction method of intelligent allocation according to claim 4 is characterized in that: A dynamic time warping algorithm is used to warp the time points of the vehicle driving state graph and the historical vehicle driving state graph in the data information sample set, and the optimal matching path of the two graphs on the time axis is found; Along the optimal matching path, the similarity of the vehicle driving state at the corresponding time point is calculated. For the calculation of the similarity of the vehicle driving state, the similarity calculation formula based on the Euclidean distance is adopted to calculate the distribution similarity of the relative angle θ, the vehicle speed V, the distance D between the vehicle and the obstacle, and the wheel steering speed V1 respectively.

6. The intelligently allocated dual-screen interaction method according to claim 5, characterized in that: The similarity range values ​​are set respectively, and weights are assigned based on the similarity range values ​​and the importance of the relative angle θ, the vehicle speed V, the distance D between the vehicle and the obstacle, and the wheel steering speed V1. A weighted total score is calculated based on the weight of each evaluation indicator and the vehicle's performance on the indicator. The weighted total score can be calculated using the following formula: Among them, Q is the weighted total score, n is the number of evaluation indicators, and f i is the score of indicator i, g i The weight of indicator i is used to obtain f i When calculating the score of a feature vector, the corresponding preset threshold interval can be set by the feature vector, and the score proportion of the feature vector falling into different preset threshold intervals can be distributed.

7. The dual-screen interaction method of intelligent allocation according to claim 6, characterized in that: Importance division: relative angle θ> vehicle speed V> distance D between vehicle and obstacle> wheel steering speed V1; IN s =In V +W d +W θ +W w ; W V , W d , W θ , W w The weights correspond to the relative angle, vehicle speed, distance between the vehicle and the obstacle, and wheel steering speed respectively, and the sum of the weights is 1. The weighted total score Q is compared with the set comprehensive evaluation threshold. If the weighted total score is higher than the threshold, it is considered that the vehicle has the conditions to pass the obstacle and the vehicle is driving normally; if it is lower than the threshold, it is considered that the vehicle does not have the conditions to pass the obstacle.

8. The intelligently allocated dual-screen interaction method according to claim 7, characterized in that: When processing the image input from the operation screen to the display screen, the point operation algorithm in the image enhancement technology is used to enhance the image in the image sample, as follows: Change the contrast of the image by linearly stretching or compressing the grayscale value of the image; h(j│k)=a×q(j│k)+b; Among them, q(j│k) is the grayscale value of the original image, h(j│k) is the grayscale value after transformation, and a and b are the coefficients of linear transformation.

9. An intelligently allocated dual-screen interactive terminal, characterized in that: It includes the interactive method as described in claim 1, in which the operation screen, the display screen and the vehicle system establish a communication connection, a right interactive module is set between the vehicle system and the display screen, and the interactive module is used to respond to the user's interactive instructions on the operation screen and map the interactive operations generated on the operation screen to the display screen.

Images