Auxiliary obstacle avoidance method and system of snowfield scooter
Through infrared detection components and millimeter wave radar, obstacle distribution point clouds are generated, combined with the Internet of Things, the three-dimensional landform model of snow is constructed, passability is analyzed and obstacle avoidance paths are planned, which solves the problem of poor obstacle avoidance effect of snow scooters and significantly improves snow safety of ground driving.
Patent Information
- Application Number
- CN202411108934.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-13
- Publication Date
- 2024-11-22
AI Technical Summary
The existing snow scooter obstacle avoidance technology mainly relies on human judgment, resulting in poor obstacle avoidance effect and inability to effectively ensure the safety of snow driving.
The infrared detection component and millimeter-wave radar generate obstacle distribution point clouds, combine the Internet of Things technology to build a three-dimensional landform model in snow, perform the overlap between the obstacle distribution point cloud and the landform model, analyze the passivity evaluation value, plan obstacle avoidance paths and send optimized steering Angle and speed warning.
The safety of snow scooters driving in snow has been improved, and the obstacle avoidance effect has been significantly improved through automated obstacle avoidance technology, reducing the dependence on human judgment.
Smart Images

Figure CN119002497A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of auxiliary obstacle avoidance, and in particular to an auxiliary obstacle avoidance method and system for a snow scooter. Background Art
[0002] Snow scooters are a type of vehicle that travels on snow. They are small, flexible, and highly adaptable, making them a popular choice for winter sports and entertainment. However, due to the complexity and uncertainty of the snowy environment, scooters often face obstacles and dangerous situations during driving. In order to improve the safety performance and user experience of snow scooters, assisted obstacle avoidance technology has become a key research area.
[0003] The obstacle avoidance work of the snow scooter in the prior art mainly relies on human judgment, resulting in poor obstacle avoidance effect, which ultimately makes it impossible to ensure the safety of the snow scooter traveling on the snow. Summary of the invention
[0004] The present application provides an auxiliary obstacle avoidance method and system for a snow scooter, which solves the problem in the prior art that the obstacle avoidance work of the snow scooter mainly relies on human judgment, resulting in poor obstacle avoidance effect, so that the safety of the snow scooter traveling on the snow is guaranteed.
[0005] In view of the above problems, the present application provides an auxiliary obstacle avoidance method for a snow scooter.
[0006] In a first aspect, the present application provides an auxiliary obstacle avoidance method for a snow scooter, the method comprising: receiving a traveling direction of a snow scooter and configuring a monitoring constraint area; activating an infrared detection component and a millimeter-wave radar to perform signal collection on the monitoring constraint area to generate an obstacle distribution point cloud, wherein the infrared detection component and the millimeter-wave radar are installed on the snow scooter; based on the Internet of Things, loading snow depth monitoring data and snow-free landform monitoring data of the monitoring constraint area to construct a three-dimensional snow landform model; overlapping the distribution coordinate system of the obstacle distribution point cloud and the three-dimensional snow landform model to generate a landform quasi-obstacle distribution point cloud and a foreign object obstacle distribution point cloud; performing safety analysis on the landform quasi-obstacle distribution point cloud to generate a passability evaluation value; setting the landform quasi-obstacle distribution point cloud whose passability evaluation value meets the passability threshold as a landform obstacle distribution point cloud; performing obstacle avoidance path planning based on the landform obstacle distribution point cloud and the foreign object obstacle distribution point cloud to generate a scooter obstacle avoidance optimization steering angle and a scooter obstacle avoidance optimization speed, and sending them to the snow scooter user end to perform obstacle avoidance warning.
[0007] In the second aspect, the present application provides an auxiliary obstacle avoidance system for a snow scooter, the system comprising: a monitoring constraint area module: receiving the direction of travel of the snow scooter and configuring a monitoring constraint area; an obstacle distribution point cloud module: activating an infrared detection component and a millimeter-wave radar to perform signal collection on the monitoring constraint area and generate an obstacle distribution point cloud, wherein the infrared detection component and the millimeter-wave radar are installed on the snow scooter; a three-dimensional terrain model module: based on the Internet of Things, loading snow depth monitoring data and snow-free terrain monitoring data of the monitoring constraint area, and constructing a three-dimensional terrain model of snow; a distribution point cloud overlap module: overlapping the obstacle distribution point cloud and The distribution coordinate system of the three-dimensional snow terrain model generates a terrain quasi-obstacle distribution point cloud and a foreign object obstacle distribution point cloud; a passability evaluation value module: performs safety analysis on the terrain quasi-obstacle distribution point cloud to generate a passability evaluation value; a terrain obstacle distribution module: sets the terrain quasi-obstacle distribution point cloud whose passability evaluation value meets the passability threshold as the terrain obstacle distribution point cloud; an obstacle avoidance optimization speed module: performs obstacle avoidance path planning based on the terrain obstacle distribution point cloud and the foreign object obstacle distribution point cloud, generates a scooter obstacle avoidance optimization steering angle and a scooter obstacle avoidance optimization speed, and sends them to the snow scooter user end to execute obstacle avoidance warning.
[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:
[0009] The auxiliary obstacle avoidance method and system for a snow scooter provided in an embodiment of the present application receives the direction of travel of the snow scooter, configures a monitoring constraint area, activates the infrared detection component and the millimeter wave radar to perform signal collection on the monitoring constraint area, generates an obstacle distribution point cloud, and then loads the snow depth monitoring data and snow-free landform monitoring data of the monitoring constraint area based on the Internet of Things, constructs a three-dimensional landform model of the snow, and overlaps the distribution coordinate system of the obstacle distribution point cloud and the three-dimensional landform model of the snow to generate a landform quasi-obstacle distribution point cloud and a foreign body obstacle distribution point cloud, and performs Safety analysis is performed to generate a passability evaluation value, and then the terrain quasi-obstacle distribution point cloud whose passability evaluation value meets the passability threshold is set as the terrain obstacle distribution point cloud. Finally, obstacle avoidance path planning is performed based on the terrain obstacle distribution point cloud and the foreign object obstacle distribution point cloud, and the scooter obstacle avoidance optimization steering angle and scooter obstacle avoidance optimization speed are generated, which are sent to the snow scooter user end to execute obstacle avoidance warning, which solves the problem of poor obstacle avoidance effect of snow scooter obstacle avoidance work in the prior art due to main reliance on human judgment, so that the safety of snow scooters driving on snow is guaranteed. BRIEF DESCRIPTION OF THE DRAWINGS
[0010] Figure 1 A schematic flow chart of an auxiliary obstacle avoidance method for a snow scooter is provided for this application;
[0011] Figure 2A schematic structural diagram of an auxiliary obstacle avoidance system for a snow scooter is provided for this application.
[0012] Explanation of the reference numerals: monitoring constraint area module 11, obstacle distribution point cloud module 12, three-dimensional terrain model module 13, distribution point cloud overlap module 14, passability evaluation value module 15, terrain obstacle distribution module 16, obstacle avoidance optimization speed module 17. DETAILED DESCRIPTION
[0013] The present application provides an auxiliary obstacle avoidance method and system for a snow scooter. The method receives the direction of travel of the snow scooter, configures a monitoring constraint area, activates an infrared detection component and a millimeter-wave radar to perform signal collection on the monitoring constraint area, generates an obstacle distribution point cloud, and then loads snow depth monitoring data and snow-free landform monitoring data of the monitoring constraint area based on the Internet of Things, constructs a three-dimensional snow landform model, and overlaps the distribution coordinate system of the obstacle distribution point cloud and the three-dimensional snow landform model to generate a landform quasi-obstacle distribution point cloud and a foreign object obstacle distribution point cloud. The landform quasi-obstacle distribution point cloud is safety analyzed to generate a passability evaluation value, and then the landform quasi-obstacle distribution point cloud whose passability evaluation value meets the passability threshold is set as a landform obstacle distribution point cloud. Finally, obstacle avoidance path planning is performed based on the landform obstacle distribution point cloud and the foreign object obstacle distribution point cloud, and a scooter obstacle avoidance optimization steering angle and a scooter obstacle avoidance optimization speed are generated, which are sent to the snow scooter user end to perform obstacle avoidance warning. The present invention solves the problem that the obstacle avoidance of the snow scooter in the prior art is poor due to the main reliance on human judgment, thereby ensuring the safety of the snow scooter traveling on the snow.
[0014] Embodiment 1
[0015] like Figure 1 As shown, the present application provides an auxiliary obstacle avoidance method and system for a snow scooter, the method comprising:
[0016] Receive the direction of travel of the snow scooter and configure the monitoring restriction area;
[0017] Loading basic information of the scooter, wherein the basic information of the scooter includes a scooter passage constraint width and a scooter passage constraint height;
[0018] configuring a monitoring distance rating, combining the scooter passage constraint width and the scooter passage constraint height, to construct a monitoring constraint space, wherein the monitoring constraint space has a scooter embedded position tag;
[0019] The scooter embedded position tag is overlapped with the scooter real-time position, and the monitoring constraint space is deployed along the travel direction of the snow scooter to generate the monitoring constraint area.
[0020] The basic information of the scooter includes the scooter's passage constraint width and the scooter's passage constraint height. The passage constraint width refers to the minimum width that the snow scooter can pass through, and the passage constraint width is not equal to the snow scooter's width. Similarly, the passage constraint height refers to the minimum height that the snow scooter can pass through, and the passage constraint height is not equal to the snow scooter's height. The height and width information of the scooter are obtained based on the basic information of the scooter, and the scooter's passage constraint width and scooter's passage constraint height are obtained based on the scooter's height and width information. The monitoring distance rating refers to the maximum distance for monitoring the scooter. The passage constraint width, scooter's passage constraint height and the detection distance rating are used as the width, height and length to construct the monitoring constraint space, and the monitoring constraint space is labeled according to the scooter's embedded position. The label is a coordinate position label. The scooter's embedded position label and the scooter's real-time position are overlapped, and the monitoring constraint space is deployed along the direction of the snow scooter's travel to complete the generation of the monitoring constraint area. By pre-configuring a rectangular space and presetting the embedding position, and then overlapping the embedding position with the scooter, the positioning data of the actual monitoring constraint area is made more accurate.
[0021] activating an infrared detection component and a millimeter-wave radar to perform signal collection on the monitoring constraint area to generate an obstacle distribution point cloud, wherein the infrared detection component and the millimeter-wave radar are installed on the snow scooter;
[0022] Obstacle detection is performed inside the snow through the infrared detection component. Since the temperature of objects on the ground inside the snow is higher than that of the snow, the obstacle distribution point cloud inside the snow can be detected, and the obstacle distribution point cloud above the snow surface is detected through the millimeter-wave radar. The signal of the monitoring constraint area is obtained through the infrared detection component and the millimeter-wave radar, and the data is parsed according to the collected signal to obtain the obstacle distribution point cloud. Point cloud is a data structure commonly used to represent 3D objects or environments, and each point represents a specific location. By installing the infrared detection component and the millimeter-wave radar on the snow scooter to obtain the obstacle distribution point cloud, it provides a basis for the subsequent loading of the snow depth monitoring data and the snow-free landform monitoring data of the monitoring constraint area based on the Internet of Things, and the construction of a three-dimensional snow landform model.
[0023] Based on the Internet of Things, the snow depth monitoring data and the snow-free landform monitoring data of the monitoring constraint area are loaded to construct a three-dimensional landform model of the snow area;
[0024] The snow depth monitoring data and snow-free landform monitoring data of the monitoring constraint area are collected through sensors, that is, real-time monitoring is carried out through the snow depth detection instruments deployed in the detection constraint area, and the data is transmitted to the system through the Internet of Things network. Similarly, the landform mapping data of the detection constraint area is obtained when there is no snow accumulation, and the snow-free landform monitoring data is obtained, and the snow-free landform monitoring data is transmitted to the system through the Internet of Things network. Further construct a three-dimensional snow model, and use a geographic information system or related three-dimensional modeling software to construct a three-dimensional landform model of the snow based on the snow depth monitoring data and the snow-free landform monitoring data. The constructed three-dimensional snow landform model indicates the corresponding snow depth on the ground, which provides a data basis for the subsequent overlap of the obstacle distribution point cloud and the distribution coordinate system of the three-dimensional snow landform model, and the generation of the landform quasi-obstacle distribution point cloud and the foreign body obstacle distribution point cloud.
[0025] Overlapping the obstacle distribution point cloud with the distribution coordinate system of the snow three-dimensional landform model to generate a landform quasi-obstacle distribution point cloud and a foreign body obstacle distribution point cloud;
[0026] Counting the distribution coordinate system of the obstacle distribution point cloud and the three-dimensional snow landform model to obtain the distribution coordinate information of the three-dimensional snow landform model and the coordinate information of the obstacle distribution point cloud;
[0027] The three-dimensional snow landform model distribution coordinate information includes snow-free landform distribution coordinate information;
[0028] When the horizontal coordinates are the same and the obstacle distribution point cloud coordinate information is below the snow-free landform distribution coordinate information, it is set as the landform quasi-obstacle distribution point cloud;
[0029] When the horizontal coordinates are the same and the obstacle distribution point cloud coordinate information does not fall below the snow-free landform distribution coordinate information, it is set as the foreign body obstacle distribution point cloud.
[0030] The quasi-obstacle distribution point cloud of topography refers to the obstacle data at the beginning of monitoring, which may be the echo point cloud caused by the uneven topography in front, that is, the obstacles existing in the topography itself, such as rocks and land protrusions. The foreign body obstacle distribution point cloud refers to the obstacles that may be non-topographic obstacles, such as man-made garbage. The coordinate data of the obstacle distribution point cloud and the three-dimensional topography model of the snow are statistically analyzed to obtain their respective distribution coordinate systems. Through the statistically obtained distribution coordinate system, the distribution coordinate information of the three-dimensional topography model of the snow and the coordinate information of the obstacle distribution point cloud are extracted, and the distribution coordinate information of the three-dimensional topography model of the snow includes the distribution coordinate information of the snow-free topography. When the horizontal coordinates are the same and the coordinate information of the obstacle distribution point cloud is below the snow-free landform distribution coordinate information, it is set as a landform quasi-obstacle distribution point cloud, that is, the same horizontal coordinates, the height is equal to or less than the snow-free landform distribution coordinates, it belongs to the landform quasi-obstacle distribution point cloud; similarly, when the horizontal coordinates are the same and the obstacle distribution point cloud coordinate information does not belong to the snow-free landform distribution coordinate information, it is set as a foreign body obstacle distribution point cloud, that is, the same horizontal coordinates, the height is greater than the snow-free landform distribution coordinates, it belongs to the foreign body obstacle distribution point cloud. The horizontal coordinates are used to judge the landform quasi-obstacle distribution point cloud and the foreign body obstacle distribution point cloud to improve the accuracy of the distribution point cloud.
[0031] Performing safety analysis on the topographic quasi-obstacle distribution point cloud to generate a passability evaluation value;
[0032] Matching a predicted travel path curve from the three-dimensional snow terrain model according to the terrain quasi-obstacle distribution point cloud and the travel direction of the snow scooter;
[0033] Extracting an upward slope change sequence, a downward slope change sequence and a total upward distance value of the travel prediction path curve;
[0034] Configure the slope threshold and the up distance threshold;
[0035] Extracting a first distance proportional coefficient of the upward slope change sequence that is greater than or equal to the slope threshold;
[0036] Extracting a second distance proportional coefficient of the descending slope change sequence that is greater than or equal to the slope threshold;
[0037] A third distance ratio coefficient of the total uplink distance value and the uplink distance threshold;
[0038] When the first distance ratio coefficient is less than a first distance ratio threshold, the second distance ratio coefficient is less than a second distance ratio threshold, and the third distance ratio coefficient is less than a third distance ratio threshold, setting the passability evaluation value to 1;
[0039] Otherwise, the traffic evaluation value is set to 0.
[0040] According to the location of the topographic quasi-obstacle distribution point cloud and the direction of the snow scooter, a prediction path curve suitable for the snow scooter is matched from the three-dimensional snow terrain model. The matched prediction path curve is analyzed to extract parameters such as the upward slope change sequence, the downward slope change sequence and the total value of the upward distance. Among them, the upward slope sequence refers to the upward slope sequence obtained by serializing and integrating the upward slopes on all predicted paths; the downward slope sequence refers to the downward slope sequence obtained by serializing and integrating the downward slopes on all predicted paths. According to the actual situation, the threshold of the upward slope change sequence and the total value of the upward distance is configured. The threshold is used to evaluate the risk degree of the path. When the sequence and the total value of the distance exceed the threshold, it indicates that there is a risk. The sequence of the upward slope change sequence greater than or equal to the slope threshold is compared with the total number of sequences to obtain the first distance ratio coefficient; the sequence of the downward slope change sequence greater than or equal to the slope threshold is compared with the total number of sequences to obtain the second distance ratio coefficient; the total value of the upward distance is compared with the upward distance threshold to obtain the third distance ratio coefficient. When the first distance ratio coefficient is less than the first distance ratio threshold, it means that the first distance ratio coefficient is qualified, and the second distance ratio coefficient is less than the second distance ratio threshold, it means that the second distance ratio coefficient is qualified, and the third distance ratio coefficient is less than the third distance ratio threshold, it means that the third distance ratio coefficient is qualified, and the passability evaluation value is set to 1, indicating that the upward slope change sequence, the downward slope change sequence and the total value of the upward distance are all qualified and can be passed safely; otherwise, if any one of them is not satisfied, the passability evaluation value is set to 0. Safety analysis is performed based on the upward slope change sequence, the downward slope change sequence and the total value of the upward distance, so that the accuracy of the passability evaluation value is improved.
[0041] The topographic quasi-obstacle distribution point cloud whose passability evaluation value meets the passability threshold is set as the topographic obstacle distribution point cloud;
[0042] Based on the Internet of Things, the snow water content monitoring data of the monitoring constraint area is loaded;
[0043] Based on the snow water content monitoring data, snow areas where the snow water content monitoring value is less than or equal to the snow water content threshold are added to the geomorphic obstacle distribution point cloud.
[0044] When a scooter travels on the snow, insufficient water content in the snow under the scooter can easily lead to landslides and create danger, so it is necessary to analyze the water content in the snow.
[0045] Through the Internet of Things technology, the snow water content information collected by sensors or other monitoring equipment deployed in the snow is obtained to obtain the snow water content monitoring data of the monitoring constraint area, which includes the snow water content. The loaded snow water content monitoring data is processed and analyzed, and the snow areas where the snow water content monitoring value is less than or equal to the snow water content threshold are added to the geomorphic obstacle distribution point cloud. The snow water content threshold refers to a specific value used to distinguish areas in the snow where the snow water content is too high or too low. This threshold can be set according to actual conditions, usually based on a comprehensive consideration of factors such as snow type, climate conditions, and season. By setting the snow water content threshold, areas with insufficient snow water content are added to the geomorphic obstacle distribution point cloud to reduce risks and ensure driving safety.
[0046] Obstacle avoidance path planning is performed based on the terrain obstacle distribution point cloud and the foreign object obstacle distribution point cloud, and a scooter obstacle avoidance optimization steering angle and a scooter obstacle avoidance optimization speed are generated, which are sent to the snow scooter user terminal to perform obstacle avoidance warning.
[0047] Appropriate path planning algorithms, such as the Dijkstra algorithm or the artificial potential field method, are used to plan one or more obstacle avoidance paths based on the distribution point cloud of terrain obstacles and the distribution point cloud of foreign obstacles. These algorithms can calculate the optimal or suboptimal path from the current position to the target position. After planning the obstacle avoidance path, the optimal steering angle and speed required by the scooter during driving are calculated by obtaining factors such as the curvature of the path, the dynamic model of the scooter, and the safety distance, so that the scooter can successfully avoid obstacles and pass through the obstacle area. The planned obstacle avoidance path, the optimized steering angle, and the scooter obstacle avoidance optimization speed are sent to the user end of the snow scooter in real time. After receiving the obstacle avoidance warning, the snow scooter user end reminds the user that there is an obstacle ahead and needs to take avoidance measures through sound, vibration or visual prompts. By sending the scooter obstacle avoidance optimization steering angle and scooter obstacle avoidance optimization speed to the user end for warning, the user can be reminded to take avoidance measures in time to reduce the risk of collision.
[0048] Furthermore, the method further comprises:
[0049] When the horizontal coordinates are the same and the obstacle distribution point cloud coordinate information is not below the snow-free landform distribution coordinate information, it is set as the initial foreign body obstacle distribution point cloud;
[0050] The three-dimensional snow landform model distribution coordinate information also includes snow surface landform distribution coordinate information;
[0051] Constructing a snow-inside constraint space according to the snow surface landform distribution coordinate information and the snow-free landform distribution coordinate information;
[0052] Performing obstacle risk assessment on the initial foreign object obstacle distribution point cloud according to the snow constraint space to generate a plurality of obstacle risk assessment values;
[0053] The initial foreign body obstacle distribution point cloud of the several obstacle risk evaluation values whose obstacle risk evaluation values are greater than or equal to the obstacle risk threshold is deleted to generate a foreign body obstacle distribution point cloud.
[0054] The foreign object obstacles in the foreign object obstacle distribution point cloud have different obstacle risks due to their depth position in the snow. Some foreign object obstacles will not affect the scooter, while some foreign object obstacles have greater obstacle risks, so they need to be discussed separately. When the horizontal coordinates are the same, and the obstacle distribution point cloud coordinate information does not belong to the snow-free landform distribution coordinate information, it is set as the initial foreign object obstacle distribution point cloud, and the initial foreign object obstacle distribution point cloud is further analyzed later. The distribution coordinate information of the snow three-dimensional landform model also includes the snow surface landform distribution coordinate information. According to the snow surface landform distribution coordinate information and the snow-free landform distribution coordinate information, the snow constraint space is constructed. That is, the landform conditions when there is no snow are compared with the landform conditions covered by snow, and the landform conditions only in the snow are obtained, and the landform conditions in the snow are spatially represented to obtain the snow constraint space. According to the constraint space in the snow, the obstacle risk of the initial foreign object obstacle distribution point cloud is evaluated, the risk of the obstacle foreign objects in the initial foreign object obstacle distribution point cloud is determined, and several obstacle risk evaluation values are generated. The initial foreign object obstacle distribution point cloud with several obstacle risk evaluation values greater than or equal to the obstacle risk threshold is deleted to generate a foreign object obstacle distribution point cloud. The initial foreign object obstacle distribution point cloud with several obstacle risk evaluation values greater than or equal to the obstacle risk threshold indicates that the initial obstacle foreign objects corresponding to the point cloud do not pose a threat to the scooter, and the corresponding point cloud can be eliminated to reduce data occupancy and improve calculation efficiency.
[0055] Furthermore, the method further comprises:
[0056] Based on a distance threshold, performing neighborhood clustering on the initial foreign object obstacle distribution point cloud to generate a plurality of initial foreign object obstacle distribution areas, wherein the plurality of initial foreign object obstacle distribution areas have a plurality of snow feature volumes;
[0057] The plurality of snow-based characteristic volumes are traversed to obtain the proportion of the snow-based volume threshold, and the plurality of obstacle risk assessment values are generated.
[0058] The distance threshold is the distance threshold between point cloud data. The distance of the initial foreign object obstacle distribution point cloud is judged based on the distance threshold. The adjacent initial foreign object obstacle distribution point clouds that meet the distance threshold are clustered to generate several initial foreign object obstacle distribution areas. The initial foreign object obstacle distribution area means that the close point clouds are regarded as the same object and set as an obstacle distribution area. The characteristic volume on the snow represents the volume of the obstacle on the snow, which will cause obstacles to the passage of the scooter. Traverse several characteristic volumes on the snow to obtain the proportion of the volume threshold on the snow. The volume threshold on the snow refers to the volume value that will cause obstacles to passage. The volume that meets the volume threshold on the snow is compared with the comprehensive volume of all characteristic volumes on the snow to obtain the proportion of the volume threshold on the snow, and obtain several obstacle risk evaluation values. The obstacle risk evaluation of the initial foreign object obstacle distribution point cloud according to the constraint space in the snow can make the obstacle risk evaluation value more practical and accurate.
[0059] Embodiment 2
[0060] Based on the same inventive concept as the auxiliary obstacle avoidance method for a snow scooter in the aforementioned embodiment, Figure 2 As shown, the present application provides an auxiliary obstacle avoidance system for a snow scooter, the system comprising:
[0061] Monitoring restricted area module 11: The monitoring restricted area module 11 is used to receive the direction of travel of the snow scooter and configure the monitoring restricted area;
[0062] Obstacle distribution point cloud module 12: the obstacle distribution point cloud module 12 is used to activate the infrared detection component and the millimeter wave radar to perform signal collection on the monitoring constraint area to generate an obstacle distribution point cloud, wherein the infrared detection component and the millimeter wave radar are installed on the snow scooter;
[0063] Three-dimensional landform model module 13: The three-dimensional landform model module 13 is used to load the snow depth monitoring data and snow-free landform monitoring data of the monitoring constraint area based on the Internet of Things to construct a three-dimensional landform model of snow;
[0064] Distribution point cloud overlap module 14: the distribution point cloud overlap module 14 is used to overlap the obstacle distribution point cloud and the distribution coordinate system of the snow three-dimensional landform model to generate a landform quasi-obstacle distribution point cloud and a foreign body obstacle distribution point cloud;
[0065] Passability evaluation value module 15: The passability evaluation value module 15 is used to perform safety analysis on the topographic quasi-obstacle distribution point cloud to generate a passability evaluation value;
[0066] Geomorphic obstacle distribution module 16: The geomorphic obstacle distribution module 16 is used to set the geomorphic quasi-obstacle distribution point cloud whose passability evaluation value meets the passability threshold as a geomorphic obstacle distribution point cloud;
[0067] Obstacle avoidance optimization speed module 17: The obstacle avoidance optimization speed module 17 is used to perform obstacle avoidance path planning based on the terrain obstacle distribution point cloud and the foreign object obstacle distribution point cloud, generate the scooter obstacle avoidance optimization steering angle and the scooter obstacle avoidance optimization speed, and send them to the snow scooter user end to execute obstacle avoidance warning.
[0068] Furthermore, the restricted area monitoring module 11 further includes:
[0069] Loading basic information of the scooter, wherein the basic information of the scooter includes a scooter passage constraint width and a scooter passage constraint height;
[0070] configuring a monitoring distance rating, combining the scooter passage constraint width and the scooter passage constraint height, to construct a monitoring constraint space, wherein the monitoring constraint space has a scooter embedded position tag;
[0071] The scooter embedded position tag is overlapped with the scooter real-time position, and the monitoring constraint space is deployed along the travel direction of the snow scooter to generate the monitoring constraint area.
[0072] Furthermore, the distribution point cloud overlap module 14 further includes:
[0073] Counting the distribution coordinate system of the obstacle distribution point cloud and the three-dimensional snow landform model to obtain the distribution coordinate information of the three-dimensional snow landform model and the coordinate information of the obstacle distribution point cloud;
[0074] The three-dimensional snow landform model distribution coordinate information includes snow-free landform distribution coordinate information;
[0075] When the horizontal coordinates are the same and the obstacle distribution point cloud coordinate information is below the snow-free landform distribution coordinate information, it is set as the landform quasi-obstacle distribution point cloud;
[0076] When the horizontal coordinates are the same and the obstacle distribution point cloud coordinate information does not fall below the snow-free landform distribution coordinate information, it is set as the foreign body obstacle distribution point cloud.
[0077] Furthermore, the distribution point cloud overlap module 14 further includes:
[0078] When the horizontal coordinates are the same and the obstacle distribution point cloud coordinate information is not below the snow-free landform distribution coordinate information, it is set as the initial foreign body obstacle distribution point cloud;
[0079] The three-dimensional snow landform model distribution coordinate information also includes snow surface landform distribution coordinate information;
[0080] Constructing a snow-inside constraint space according to the snow surface landform distribution coordinate information and the snow-free landform distribution coordinate information;
[0081] Performing obstacle risk assessment on the initial foreign object obstacle distribution point cloud according to the snow constraint space to generate a plurality of obstacle risk assessment values;
[0082] The initial foreign body obstacle distribution point cloud of the several obstacle risk evaluation values whose obstacle risk evaluation values are greater than or equal to the obstacle risk threshold is deleted to generate a foreign body obstacle distribution point cloud.
[0083] Furthermore, the distribution point cloud overlap module 14 further includes:
[0084] Based on a distance threshold, performing neighborhood clustering on the initial foreign object obstacle distribution point cloud to generate a plurality of initial foreign object obstacle distribution areas, wherein the plurality of initial foreign object obstacle distribution areas have a plurality of snow feature volumes;
[0085] The plurality of snow-based characteristic volumes are traversed to obtain the proportion of the snow-based volume threshold, and the plurality of obstacle risk assessment values are generated.
[0086] Furthermore, the traffic evaluation value module 15 also includes:
[0087] Matching a predicted travel path curve from the three-dimensional snow terrain model according to the terrain quasi-obstacle distribution point cloud and the travel direction of the snow scooter;
[0088] Extracting an upward slope change sequence, a downward slope change sequence and a total upward distance value of the travel prediction path curve;
[0089] Configure the slope threshold and the up distance threshold;
[0090] Extracting a first distance proportional coefficient of the upward slope change sequence that is greater than or equal to the slope threshold;
[0091] Extracting a second distance proportional coefficient of the descending slope change sequence that is greater than or equal to the slope threshold;
[0092] A third distance ratio coefficient of the total uplink distance value and the uplink distance threshold;
[0093] When the first distance ratio coefficient is less than a first distance ratio threshold, the second distance ratio coefficient is less than a second distance ratio threshold, and the third distance ratio coefficient is less than a third distance ratio threshold, setting the passability evaluation value to 1;
[0094] Otherwise, the traffic evaluation value is set to 0.
[0095] Furthermore, the terrain obstacle distribution module 16 also includes:
[0096] Based on the Internet of Things, the snow water content monitoring data of the monitoring constraint area is loaded;
[0097] Based on the snow water content monitoring data, snow areas where the snow water content monitoring value is less than or equal to the snow water content threshold are added to the geomorphic obstacle distribution point cloud.
[0098] Through the above detailed description of the auxiliary obstacle avoidance method for a snow scooter in this specification, those skilled in the art can clearly understand the auxiliary obstacle avoidance method for a snow scooter in this embodiment. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.
[0099] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A method for assisting obstacle avoidance of a snow scooter, characterized in that the method include: Receive the direction of travel of the snow scooter and configure the monitoring restriction area; activating an infrared detection component and a millimeter-wave radar to perform signal collection on the monitoring constraint area to generate an obstacle distribution point cloud, wherein the infrared detection component and the millimeter-wave radar are installed on the snow scooter; Based on the Internet of Things, the snow depth monitoring data and the snow-free landform monitoring data of the monitoring constraint area are loaded to construct a three-dimensional landform model of the snow area; Overlapping the obstacle distribution point cloud with the distribution coordinate system of the snow three-dimensional landform model to generate a landform quasi-obstacle distribution point cloud and a foreign body obstacle distribution point cloud; Performing safety analysis on the topographic quasi-obstacle distribution point cloud to generate a passability evaluation value; The topographic quasi-obstacle distribution point cloud whose passability evaluation value meets the passability threshold is set as the topographic obstacle distribution point cloud; Obstacle avoidance path planning is performed based on the terrain obstacle distribution point cloud and the foreign object obstacle distribution point cloud, and a scooter obstacle avoidance optimization steering angle and a scooter obstacle avoidance optimization speed are generated, which are sent to the snow scooter user terminal to perform obstacle avoidance warning.
2. The method according to claim 1, characterized in that Receive the direction of the snow scooter and configure the monitoring restriction area, including: Loading basic information of the scooter, wherein the basic information of the scooter includes a scooter passage constraint width and a scooter passage constraint height; configuring a monitoring distance rating, combining the scooter passage constraint width and the scooter passage constraint height, to construct a monitoring constraint space, wherein the monitoring constraint space has a scooter embedded position tag; The scooter embedded position tag is overlapped with the scooter real-time position, and the monitoring constraint space is deployed along the travel direction of the snow scooter to generate the monitoring constraint area.
3. The method according to claim 1, characterized in that Overlapping the obstacle distribution point cloud and the distribution coordinate system of the snow three-dimensional landform model to generate a landform quasi-obstacle distribution point cloud and a foreign body obstacle distribution point cloud, including: Counting the distribution coordinate system of the obstacle distribution point cloud and the three-dimensional snow landform model to obtain the distribution coordinate information of the three-dimensional snow landform model and the coordinate information of the obstacle distribution point cloud; The three-dimensional snow landform model distribution coordinate information includes snow-free landform distribution coordinate information; When the horizontal coordinates are the same and the obstacle distribution point cloud coordinate information is below the snow-free landform distribution coordinate information, it is set as the landform quasi-obstacle distribution point cloud; When the horizontal coordinates are the same and the obstacle distribution point cloud coordinate information does not fall below the snow-free landform distribution coordinate information, it is set as the foreign body obstacle distribution point cloud.
4. The method according to claim 3, characterized in that When the horizontal coordinates are the same and the obstacle distribution point cloud coordinate information is not below the snow-free landform distribution coordinate information, it is set as the foreign body obstacle distribution point cloud, including: When the horizontal coordinates are the same and the obstacle distribution point cloud coordinate information is not below the snow-free landform distribution coordinate information, it is set as the initial foreign body obstacle distribution point cloud; The three-dimensional snow landform model distribution coordinate information also includes snow surface landform distribution coordinate information; Constructing a snow-inside constraint space according to the snow surface landform distribution coordinate information and the snow-free landform distribution coordinate information; Performing obstacle risk assessment on the initial foreign object obstacle distribution point cloud according to the snow constraint space to generate a plurality of obstacle risk assessment values; The initial foreign body obstacle distribution point cloud of the several obstacle risk evaluation values whose obstacle risk evaluation values are greater than or equal to the obstacle risk threshold is deleted to generate a foreign body obstacle distribution point cloud.
5. The method according to claim 4, characterized in that The obstacle risk evaluation is performed on the initial foreign object obstacle distribution point cloud according to the snow constraint space to generate a plurality of obstacle risk evaluation values, including: Based on a distance threshold, performing neighborhood clustering on the initial foreign object obstacle distribution point cloud to generate a plurality of initial foreign object obstacle distribution areas, wherein the plurality of initial foreign object obstacle distribution areas have a plurality of snow feature volumes; The plurality of snow-based characteristic volumes are traversed to obtain the proportion of the snow-based volume threshold, and the plurality of obstacle risk assessment values are generated.
6. The method according to claim 1, characterized in that The safety analysis of the topographic quasi-obstacle distribution point cloud is performed to generate a passability evaluation value, including: Matching a predicted travel path curve from the three-dimensional snow terrain model according to the terrain quasi-obstacle distribution point cloud and the travel direction of the snow scooter; Extracting an upward slope change sequence, a downward slope change sequence and a total upward distance value of the travel prediction path curve; Configure the slope threshold and the up distance threshold; Extracting a first distance proportional coefficient of the upward slope change sequence that is greater than or equal to the slope threshold; Extracting a second distance proportional coefficient of the descending slope change sequence that is greater than or equal to the slope threshold; A third distance ratio coefficient of the total uplink distance value and the uplink distance threshold; When the first distance ratio coefficient is less than a first distance ratio threshold, the second distance ratio coefficient is less than a second distance ratio threshold, and the third distance ratio coefficient is less than a third distance ratio threshold, setting the passability evaluation value to 1; Otherwise, the traffic evaluation value is set to 0.
7. The method according to claim 1, characterized in that The geomorphic quasi-obstacle distribution point cloud whose passability evaluation value satisfies the passability threshold is set as a geomorphic obstacle distribution point cloud, further comprising: Based on the Internet of Things, the snow water content monitoring data of the monitoring constraint area is loaded; based on the snow water content monitoring data, the snow area where the snow water content monitoring value is less than or equal to the snow water content threshold is added to the geomorphic obstacle distribution point cloud.
8. The auxiliary obstacle avoidance system of the snow scooter is characterized by: The system comprises: Monitoring restricted area module: receives the direction of the snow scooter and configures the monitoring restricted area; Obstacle distribution point cloud module: activating the infrared detection component and the millimeter wave radar to perform signal collection on the monitoring constraint area to generate an obstacle distribution point cloud, wherein the infrared detection component and the millimeter wave radar are installed on the snow scooter; Three-dimensional landform model module: Based on the Internet of Things, the snow depth monitoring data and snow-free landform monitoring data of the monitoring constraint area are loaded to construct a three-dimensional landform model of snow; Distribution point cloud overlap module: overlaps the obstacle distribution point cloud and the distribution coordinate system of the snow three-dimensional landform model to generate landform quasi-obstacle distribution point cloud and foreign body obstacle distribution point cloud; Passability evaluation value module: performs safety analysis on the topographic quasi-obstacle distribution point cloud to generate a passability evaluation value; Geomorphic obstacle distribution module: setting the geomorphic quasi-obstacle distribution point cloud whose passability evaluation value meets the passability threshold as the geomorphic obstacle distribution point cloud; Obstacle avoidance optimization speed module: performs obstacle avoidance path planning based on the terrain obstacle distribution point cloud and the foreign object obstacle distribution point cloud, generates the scooter obstacle avoidance optimization steering angle and the scooter obstacle avoidance optimization speed, and sends them to the snow scooter user end to execute obstacle avoidance warning.
