Intelligent cone barrel automatic arming and remote cooperative control method and system

Abstract

The application provides a smart cone barrel automatic deployment and remote cooperative control method and system, relates to the technical field of intelligent traffic control, and comprises the following steps: acquiring road characteristics and task parameters, and establishing a rule model to calculate a target deployment position. After the initial trajectory of each cone barrel is planned, time-space conflict detection is performed by a cooperative node, a prediction matrix is constructed, and then the trajectory is re-planned through an optimization algorithm, different path points are distributed or the speed is adjusted to generate a conflict-free cooperative scheme. The cone barrel executes movement and reports the position, can respond to the change request of a remote terminal in real time, generates a new position sequence, and issues an instruction. The application realizes the automation, cooperation and remote controllability of the cone barrel deployment, and improves the deployment efficiency and safety.

Description

Technical Field

[0001] This invention relates to the field of intelligent traffic control technology, and in particular to a method and system for automatic deployment and remote collaborative control of intelligent traffic cones. Background Technology

[0002] In scenarios such as road construction, accident handling, or temporary traffic control, it is often necessary to quickly and accurately deploy a large number of traffic cones to isolate dangerous areas and guide traffic flow. Currently, cone deployment primarily relies on manual operation. Workers place each cone manually based on experience, according to the road conditions and deployment requirements. For tasks requiring dynamic adjustments to the deployment area or the shape of the cone array, manual handling and repositioning of each cone are also necessary. Furthermore, during deployment, workers frequently navigate through traffic, posing significant safety risks. A slightly improved approach is to use individual intelligent cones with self-moving capabilities, but their action decisions are often isolated. Each cone independently plans its path and moves based solely on its received final target location, lacking coordination and information exchange with other cones.

[0003] Furthermore, manual deployment is inefficient, slow to respond, and ill-suited for complex deployment tasks requiring rapid deployment or frequent adjustments, posing a continuous threat to the safety of personnel. While isolated intelligent cones reduce human intervention, scenarios where multiple cones move simultaneously towards the target area are prone to path intersections and conflicts. Due to the lack of global coordination and conflict prediction mechanisms, cones may collide, become blocked, or form deadlocks in narrow sections, causing delays or even failures in the overall deployment task, thus failing to achieve efficient and orderly automated deployment operations. Summary of the Invention

[0004] This invention provides a method and system for automatic deployment and remote collaborative control of intelligent cones, which can solve the problems in the prior art.

[0005] A first aspect of the present invention provides a method for automatic deployment and remote collaborative control of intelligent cone-shaped defenses, comprising: Obtain road feature information and deployment task parameters for the area to be defended; Based on the road feature information and deployment task parameters, location adaptation rules are established and a spatial constraint model is constructed. The target deployment positions of each smart cone are calculated and assigned to the corresponding cone units. Each cone unit plans an initial movement trajectory based on its current position and the target deployment position. The initial movement trajectory is then sent to the collaborative planning node via a wireless communication module for spatiotemporal conflict detection. This process identifies overlapping trajectory segments and collision times, and constructs a path conflict prediction matrix. Based on the path conflict prediction matrix, the movement trajectories of conflicting cone units are replanned using a cooperative trajectory optimization algorithm. Different intermediate path points are assigned to the conflicting cone units or their respective movement speeds are adjusted to generate a conflict-free cooperative movement scheme. Each cone unit moves according to the cooperative movement scheme and reports its location information. It receives deployment change requests from the remote control terminal in real time, parses the target area parameters in the deployment change request, generates a new cone deployment position sequence, and issues movement commands.

[0006] Based on the road feature information and deployment task parameters, location adaptation rules are established and a spatial constraint model is constructed. The target deployment positions of each smart cone are calculated and assigned to the corresponding cone units, including: The number of lanes, shoulder width, longitudinal profile data and real-time traffic data are extracted from the road feature information. The cone placement level and lateral safety distance of the cones are determined according to the number of lanes and shoulder width. The longitudinal spacing of the cones is adjusted according to the real-time traffic data to obtain the position adaptation rules. The construction impact range and traffic organization plan are determined based on the deployment task parameters. The traffic organization plan includes lane closure methods and diversion paths. Combined with the longitudinal profile alignment data, a spatial constraint model for dynamic traffic flow guidance is constructed. The spatial constraint model includes cone gradient alignment constraints based on diversion paths, lateral position offset constraints based on lane closure methods, and cone sight distance guarantee constraints based on longitudinal profile slope. The constraint parameters of the spatial constraint model are calibrated using the real-time traffic data. The location adaptation rules are coupled with the spatial constraint model to generate a target deployment location sequence. Based on the current position of each cone unit, a bilateral matching iteration is used to allocate the location, and the allocation result is sent to the corresponding cone unit.

[0007] The location adaptation rules are coupled with the spatial constraint model to generate a target deployment location sequence. Based on the current position of each cone unit, a bilateral matching iteration is used for location allocation. The allocation results are then sent to the corresponding cone unit, including: The location adaptation rules are substituted into the constraints of the spatial constraint model for coupled calculation. Within the spatial range of the constraints, a set of candidate deployment locations is generated according to the location adaptation rules, and spatial continuity is checked. Transition location points are supplemented by interpolation to generate a sequence of target deployment locations. Obtain the current position coordinates and remaining power of each cone unit, calculate the travel time and energy consumption of each cone unit to reach each position point in the target deployment position sequence, generate a cone unit preference ranking table for each cone unit to the target deployment position sequence, and a target position preference ranking table for each target deployment position to the cone unit. Based on the bucket unit preference ranking table and the target location preference ranking table, a bilateral matching iteration is performed. Each bucket unit initiates a matching request to the target deployment position with the highest ranking in the bucket unit preference ranking table. Each target deployment position temporarily accepts the highest-ranked bucket unit and rejects other bucket units according to the target location preference ranking table. The rejected bucket unit initiates a new matching request to the second-highest ranked target deployment position. This iteration is repeated until all bucket units are accepted, and a location allocation result is generated and sent to the corresponding bucket unit.

[0008] Each cone unit plans an initial movement trajectory based on its current position and the target deployment position. This initial movement trajectory is then transmitted to the collaborative planning node via a wireless communication module for spatiotemporal conflict detection. This process identifies overlapping trajectory segments and their collision times, and constructs a path conflict prediction matrix, including: An initial movement trajectory is planned based on the current position and the target deployment position. A trajectory envelope is constructed by combining the physical radius of the cone. The trajectory envelope is the boundary of a buffer zone with the initial movement trajectory as the center line and the physical radius of the cone as the offset distance. The trajectory envelope and the moving speed are sent to the collaborative planning node through the wireless communication module. The spatial intersection of the trajectory envelopes of each cone unit is determined pairwise. The overlapping area of ​​the paired trajectory envelopes is calculated, and the boundary coordinates of the overlapping area are extracted as the intersecting trajectory segments. Calculate the entry time of each cone unit entering the overlapping region and the exit time of each cone unit leaving the overlapping region. Determine whether there is a time overlap between the entry and exit times of paired cone units. When the entry time of one cone unit is earlier than the exit time of another cone unit but later than the entry time of another cone unit, record the start time of the time overlap interval as the collision time. Construct a path conflict prediction matrix. The row index and column index of the path conflict prediction matrix correspond to each cone unit. The matrix elements store the intersection and overlap trajectory segments and collision times of paired cone units.

[0009] Based on the path conflict prediction matrix, the movement trajectories of conflicting cone units are replanned using a cooperative trajectory optimization algorithm. Different intermediate path points are assigned to the conflicting cone units, or their respective movement speeds are adjusted, generating a conflict-free cooperative movement scheme, including: Extract the conflicting cone units, overlapping trajectory segments, and collision times from the path conflict prediction matrix; construct a virtual gravitational field using the target deployment position of each cone unit as the gravitational source, and the virtual gravitational field generates a gravitational vector pointing to the target deployment position on the cone units; A virtual repulsive field is constructed using the overlapping trajectory segments of conflicting cone units as the repulsive force source. The virtual repulsive field generates a repulsive force vector on other cone units that deviates from the overlapping trajectory segments. The virtual gravitational field and the virtual repulsive field are superimposed and synthesized to calculate the resultant force vector of each cone unit at the current position. The moving direction and moving speed of each cone unit are adjusted according to the resultant force vector. A bypass intermediate path point is generated at a position perpendicular to the repulsive force vector. The distance from the bypass intermediate path point to the boundary of the intersecting and overlapping trajectory segment is greater than the safety distance threshold. The movement trajectory of the cone unit is updated to an adjusted trajectory that passes through the current position, bypasses the intermediate path point, and the target deployment position in sequence; the potential field calculation and motion parameter update are performed iteratively, and the resultant force vector, movement speed, and bypass intermediate path point are recalculated in each iteration until all cone units reach the target deployment position, generating a conflict-free cooperative movement scheme.

[0010] Each cone unit moves according to the aforementioned coordinated movement scheme and reports its location information. It receives deployment change requests from the remote control terminal in real time, parses the target area parameters in the deployment change request, generates a new cone deployment position sequence, and issues movement commands, including: During the movement of each cone unit, operating status parameters are collected and sent to the remote control terminal. The operating status parameters include the current position coordinates, remaining power percentage, and remaining distance to the target deployment position. The remote control terminal calculates the energy adequacy index of each cone unit based on the remaining power percentage and remaining distance. The remote control terminal receives a deployment change request and extracts the new target area boundary coordinates, generates a new cone deployment position sequence, calculates the estimated mobile energy consumption from the current position coordinates of each cone unit to each candidate position in the new cone deployment position sequence, and constructs a task allocation cost matrix. The row index of the task allocation cost matrix is ​​the cone unit identifier, the column index is the position number in the new cone deployment position sequence, and the matrix element value is the weighted sum of the estimated mobile energy consumption and the energy adequacy index. Based on the task allocation cost matrix, perform optimal matching calculation, allocate a new target placement position that minimizes the global cost to each cone unit, generate a reassignment instruction containing the cone unit identifier and the coordinates of the new target placement position, and issue the reassignment instruction to each cone unit to trigger each cone unit to update the target placement position and re-execute path planning.

[0011] A second aspect of the present invention provides an intelligent cone-shaped automatic deployment and remote collaborative control system, comprising: The information acquisition unit is used to acquire road feature information and deployment task parameters of the area to be defended. The location calculation unit is used to establish location adaptation rules and construct a spatial constraint model based on the road feature information and deployment task parameters, calculate the target deployment position of each smart cone and assign it to the corresponding cone unit. The conflict detection unit is used by each cone unit to plan an initial movement trajectory based on its current position and the target deployment position, and to send the initial movement trajectory to the collaborative planning node through the wireless communication module for spatiotemporal conflict detection, identify the trajectory segments that intersect and overlap and the collision time, and construct a path conflict prediction matrix. The trajectory optimization unit is used to replan the movement trajectory of conflicting cone units according to the path conflict prediction matrix and through a cooperative trajectory optimization algorithm, assign different intermediate path points to the conflicting cone units or adjust their respective movement speeds to generate a conflict-free cooperative movement scheme. The collaborative execution unit is used for each cone unit to move according to the collaborative movement scheme and report its position information, receive deployment change requests sent by the remote control terminal in real time, parse the target area parameters in the deployment change request, generate a new cone deployment position sequence and issue movement instructions.

[0012] A third aspect of the present invention provides an electronic device, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0013] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0014] This method enables automated and collaborative deployment of intelligent traffic cones, significantly improving operational efficiency and safety. By acquiring road features and task parameters, and establishing location adaptation rules and spatial constraint models, the optimal target deployment position for each cone can be calculated. This process ensures a high degree of matching between the cone array and road geometry, traffic flow conditions, and specific task requirements, avoiding common problems in traditional manual deployment such as uneven spacing and improper placement, thus guaranteeing the scientific validity and compliance of the deployment plan from the outset.

[0015] Using a path conflict prediction matrix, the system dynamically replans conflicting cone trajectories through a cooperative trajectory optimization algorithm. The algorithm achieves staggered movement and avoidance in both time and space by assigning differentiated intermediate pathpoints to conflicting cones or precisely adjusting their respective movement speeds, thereby generating a globally conflict-free cooperative movement scheme.

[0016] While executing the collaborative movement plan, the system supports remote real-time monitoring and dynamic adjustment. Each cone unit continuously reports its own position, and the remote control terminal can issue anti-change requests at any time. It can quickly parse new target area parameters, regenerate the cone deployment position sequence in real time, and immediately issue movement commands. This remote collaborative control capability gives the deployment operation extremely high flexibility and response speed, enabling it to quickly adapt to sudden scenarios such as traffic accident handling and construction plan changes, and achieve rapid reconstruction of the deployment formation, greatly improving the efficiency of emergency response and task management. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the intelligent cone-shaped automatic deployment and remote collaborative control method according to an embodiment of the present invention. Figure 2 This is a flowchart illustrating the method for constructing a path conflict prediction matrix according to an embodiment of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0020] Figure 1 This is a flowchart illustrating the automatic deployment and remote collaborative control method for intelligent cone-shaped defenses according to an embodiment of the present invention. Figure 1 As shown, the method includes: Obtain road feature information and deployment task parameters for the area to be defended; Based on the road feature information and deployment task parameters, location adaptation rules are established and a spatial constraint model is constructed. The target deployment positions of each smart cone are calculated and assigned to the corresponding cone units. Each cone unit plans an initial movement trajectory based on its current position and the target deployment position. The initial movement trajectory is then sent to the collaborative planning node via a wireless communication module for spatiotemporal conflict detection. This process identifies overlapping trajectory segments and collision times, and constructs a path conflict prediction matrix. Based on the path conflict prediction matrix, the movement trajectories of conflicting cone units are replanned using a cooperative trajectory optimization algorithm. Different intermediate path points are assigned to the conflicting cone units or their respective movement speeds are adjusted to generate a conflict-free cooperative movement scheme. Each cone unit moves according to the cooperative movement scheme and reports its location information. It receives deployment change requests from the remote control terminal in real time, parses the target area parameters in the deployment change request, generates a new cone deployment position sequence, and issues movement commands.

[0021] In one optional embodiment, when acquiring road feature information of the area to be deployed, the target road segment is scanned in real time by an intelligent camera and radar system controlling safety posts. The intelligent camera collects visual feature data such as lane line positions, shoulder widths, and emergency lane boundary coordinates, while simultaneously recording traffic density and vehicle speed. The radar system detects the distribution of obstacles within the emergency lane at a frequency of 10 times per second and measures the available space distance between the guardrail and the lane dividing line. The edge computing unit matches and fuses the image data with a pre-stored 3D road model, which stores precise geometric information of the road and positioning references under a unified coordinate system. By matching image feature points with landmarks in the 3D road model, the edge computing unit establishes a transformation relationship between the pixel coordinate system and the world coordinate system, thereby extracting the 3D coordinate sequence, radius of curvature, and longitudinal slope value of the lane centerline in the world coordinate system. Deployment task parameters are obtained from the instruction data packet issued by the control center, including information such as deployment type identifier, control duration, priority level, and special vehicle passage requirements.

[0022] When establishing location adaptation rules, the longitudinal baseline for cone placement is determined based on lane width and emergency lane boundary coordinates. This baseline maintains a 0.3-meter safety gap from the outer edge of the emergency lane. The longitudinal spacing rules are dynamically adjusted according to vehicle speed. When the average speed of mainline traffic reaches 100 km / h, the cone spacing is set to 15 meters; if the speed drops below 60 km / h, the spacing is shortened to 10 meters. For curved sections, the blind spot is calculated based on the radius of curvature, and the first cone is placed 50 meters before the curve entrance as a warning sign. The spatial constraint model comprehensively considers physical obstacles, traffic flow characteristics, and cone movement capabilities. When the model calibrates prohibited placement areas, it reads the precise location information of fixed facilities such as bridge expansion joints, drainage ditch covers, and communication equipment bases in the world coordinate system from the pre-stored 3D road model. Intelligent cameras are used to detect disabled vehicles or temporary objects parked in the emergency lane in real time. Visual positioning technology is combined with a 3D road model to input the world coordinates of dynamic obstacles into the constraints. The cone four-wheel chassis has a movement limit of a minimum turning radius of 2 meters. The model is set to ensure that there is a passable path between any two adjacent target placement positions, and the curvature of the path does not exceed the chassis steering capacity.

[0023] When calculating the target deployment location of each smart cone, the edge computing unit divides the road segment to be deployed into equal parts according to the number of cones, based on the coordinate sequence of the road centerline in the 3D road model. The center point of each segment serves as the initial target point. These target points are all represented by 3D coordinates in the world coordinate system. The initial target points are fine-tuned according to the position adaptation rules, and it is checked whether the point falls within the prohibited area. The spatial constraint model evaluates the line-of-sight connectivity between each target point to ensure that there is no visual obstruction between adjacent cones. After completing the position calculation, target position information containing latitude and longitude coordinates, distance from the shoulder, and orientation angle is generated for each cone and distributed to the corresponding cone unit in the form of encrypted data frames through the wireless communication module.

[0024] After receiving the target deployment location information, the cone unit activates the geolocation module to obtain its current coordinates. The geolocation module uses BeiDou satellite positioning, achieving centimeter-level accuracy, and the positioning results are represented in the world coordinate system. The edge computing unit calculates the straight-line distance and azimuth angle between the current position and the target position. The trajectory planning algorithm generates a straight-line path connecting the start and end points, and corrects the path based on the obstacle distribution map provided by the radar detection unit. When there is an obstacle on the straight-line path, the algorithm inserts a detour path point in front of the obstacle, maintaining a safety gap of at least 0.5 meters between the detour path and the obstacle edge.

[0025] After the initial movement trajectory is generated, the cone unit encapsulates the trajectory data into a message frame, containing the cone unit's unique identifier, the parameterized expression of the trajectory curve, the estimated movement time, and the timestamp sequence of critical path points. The wireless communication module operates in the 5.8 GHz band and establishes a communication link with the collaborative planning node using time division multiple access. The collaborative planning node is deployed in the edge computing unit of the control safety stake. After receiving the trajectory data from all cone units, it reconstructs the movement process of each cone in chronological order.

[0026] Spatiotemporal conflict detection determines whether there is any intersection or overlap by traversing all cone trajectory pairs. The detection algorithm discretizes each trajectory into position snapshots along the time dimension, with a time interval set to 0.1 seconds. For any two cone units, the algorithm extracts their position coordinates at the same moment and calculates the Euclidean distance. When the distance is less than the cone's chassis diameter plus a safety margin, a collision risk is determined at that moment, and the collision time and position coordinates are recorded and stored as a conflict event. The path conflict prediction matrix is ​​a square matrix with a dimension equal to the number of cones. The matrix elements record the degree of conflict between any two cones, represented by quantitative indicators such as the predicted number of collisions, minimum distance, and temporal overlap. The collaborative planning nodes identify the cone pairs with the most severe conflicts based on the distribution of non-zero elements in the matrix and prioritize trajectory replanning.

[0027] The cooperative trajectory optimization algorithm employs a hierarchical strategy to handle conflicts of varying degrees. For minor conflicts, the algorithm resolves the conflict by adjusting the movement speed and allowing one of the cones to start earlier or later, thus staggering the time it takes for both cones to pass through the conflict zone. For severe conflicts with intersecting trajectories and high time overlap, the algorithm replans a detour path for one of the cones, with the intermediate path point of the detour path chosen to be 3 to 5 meters to the side of the original trajectory intersection point. The optimization process uses an iterative mechanism, re-performing conflict detection after each adjustment until all off-diagonal elements in the matrix are zero.

[0028] When generating a conflict-free cooperative movement plan, the cooperative planning node outputs detailed execution instructions for each cone unit, including corrected trajectory parameters, segmented speed settings, and arrival time requirements for critical path points. The cooperative movement plan also includes contingency plans, specifying how adjacent cones should adjust their positions to fill the gap if a cone malfunctions and cannot move as planned. During the cone unit's movement according to the cooperative movement plan, the four-wheel motion control unit continuously monitors the actual chassis speed and steering angle. The control unit employs a closed-loop feedback mechanism, comparing the actual driving trajectory with the planned trajectory in real time, and automatically correcting the steering angle when the lateral deviation exceeds 0.1 meters. The geolocation module outputs the current position coordinates every second, and the edge computing unit reports the location information to the cooperative planning node.

[0029] When a remote control terminal sends a deployment change request to a collaborative planning node, the request message includes a change type identifier and target area parameters. Change types are categorized into three types: partial adjustment, complete evacuation, and emergency avoidance. When parsing the deployment change request, the collaborative planning node extracts the change type identifier and invokes the corresponding processing flow based on the identifier. For partial adjustment requests, the processing flow reads the coordinate range from the target area parameters and determines which cones need to be moved based on the current deployment status. For emergency avoidance requests, the processing flow calculates the estimated arrival time and travel path of whitelisted vehicles and selects the cones closest to the path for lateral movement.

[0030] The generation of new cone deployment position sequences uses the same position adaptation rules and spatial constraint model as the initial deployment, but the calculation starting point becomes the current real-time position of each cone. After generating the position sequence, the edge computing unit immediately executes a cooperative trajectory optimization algorithm to ensure no conflicts occur between cones during the change process. When issuing a movement command, the cooperative planning node sends an encrypted command frame to the target cone unit via a wireless communication module, containing the new target position coordinates, movement priority, execution time limit, and cooperative constraints. After receiving the command, the cone unit immediately plans its movement trajectory and begins execution, simultaneously activating the LED display and warning flashers to continuously warn passing vehicles to avoid it during the movement.

[0031] In one optional implementation, based on the road feature information and deployment task parameters, a location adaptation rule is established and a spatial constraint model is constructed to calculate the target deployment location of each smart cone and assign it to the corresponding cone unit, including: The number of lanes, shoulder width, longitudinal profile data and real-time traffic data are extracted from the road feature information. The cone placement level and lateral safety distance of the cones are determined according to the number of lanes and shoulder width. The longitudinal spacing of the cones is adjusted according to the real-time traffic data to obtain the position adaptation rules. The construction impact range and traffic organization plan are determined based on the deployment task parameters. The traffic organization plan includes lane closure methods and diversion paths. Combined with the longitudinal profile alignment data, a spatial constraint model for dynamic traffic flow guidance is constructed. The spatial constraint model includes cone gradient alignment constraints based on diversion paths, lateral position offset constraints based on lane closure methods, and cone sight distance guarantee constraints based on longitudinal profile slope. The constraint parameters of the spatial constraint model are calibrated using the real-time traffic data. The location adaptation rules are coupled with the spatial constraint model to generate a target deployment location sequence. Based on the current position of each cone unit, a bilateral matching iteration is used to allocate the location, and the allocation result is sent to the corresponding cone unit.

[0032] Road feature information is collected and extracted, including the number of lanes, shoulder width, longitudinal profile data, and real-time traffic data. The number of lanes is obtained by identifying lane lines using high-definition cameras or by measuring distances using lidar. Shoulder width is obtained through a road geographic information system or determined by on-site measurement. Longitudinal profile data includes information such as road slope and curvature, which can be obtained through a road management system or measured in real time by vehicle-mounted sensors. Real-time traffic data includes traffic flow, average vehicle speed, and vehicle type distribution, which are obtained through a road monitoring system or temporarily deployed traffic monitoring equipment.

[0033] The cone placement levels and lateral safety distances are determined based on the extracted number of lanes and shoulder width. Placement levels refer to the number of cone layers on the road cross-section, typically divided into three levels: mainline placement, transition zone placement, and buffer zone placement. For a two-way four-lane road, when one lane needs to be closed, mainline cones are placed along the lane edge, transition zone cones extend at a certain angle, and buffer zones form a protective strip in front of the construction area. The determination of lateral safety distances follows these rules: when the lane width is 3.5 meters, the cones should be at least 0.5 meters from the lane edge; when the shoulder width is greater than 1.5 meters, warning cones can be placed on the shoulder at a lateral distance of 1 meter; for highways, the lateral safety distance should be increased by at least 1 meter. The longitudinal spacing of the cones is adjusted based on real-time traffic data to form a complete set of location adaptation rules. When the average vehicle speed is below 40 km / h, the longitudinal spacing between cones is set to 5 to 8 meters; when the vehicle speed is between 40 and 80 km / h, the spacing is adjusted to 8 to 15 meters; and when the vehicle speed exceeds 80 km / h, the spacing is increased to 15 to 25 meters. Meanwhile, in sections with high traffic volume, the cone spacing is appropriately reduced to enhance visual guidance.

[0034] The scope of the construction impact and traffic management plan are determined based on the deployment task parameters, including the construction type, location, duration, and safety level. The scope of the construction impact is determined by the location and type of construction, such as pipeline repair, road paving, or bridge maintenance, each with a different impact range. The traffic management plan is generated based on the scope of the construction impact and includes lane closure methods and diversion paths. Lane closure methods include single-side closure, center closure, or full closure; diversion paths design routes for vehicles to bypass the construction area, including three continuous areas: a speed reduction warning zone, a transition zone, and the construction zone itself.

[0035] Based on longitudinal profile data, a spatial constraint model for dynamic traffic flow guidance is constructed, comprising three core constraints: cone gradient alignment constraints based on the guidance path, lateral position offset constraints based on lane closure methods, and cone sight distance assurance constraints based on longitudinal profile slope. For cone gradient alignment constraints, cones are arranged diagonally in lane transition areas to guide vehicles smoothly from the original lane to the temporary lane. The length of the gradient alignment is proportional to the design speed and generally adopts a gentle curve design to ensure a smooth change in vehicle steering angle. For roads with a design speed of 60 km / h, the gradient zone length is no less than 60 meters; for speeds of 80 km / h, the gradient zone length should reach at least 90 meters.

[0036] Lateral position offset constraints determine the precise position of the cones on the road cross section based on the lane closure method. When the right lane is closed, the cones are laid out starting from the right shoulder and gradually shifted to the left to the left edge of the closed lane. When the left lane is closed, the cones are laid out starting from the central divider and shifted to the right to the right edge of the closed lane. For the closure of the middle lane, a bidirectional offset cone layout needs to be designed to form a "funnel-shaped" diversion area.

[0037] The cone visibility constraint takes into account the impact of road longitudinal profile slope on cone visibility. On uphill sections, due to limited visibility, the cone spacing needs to be reduced to enhance visual guidance. On downhill sections with a slope greater than 3%, the cone spacing is reduced by 20% compared to flat sections, and warning cones are added in front of the top of the slope. During nighttime construction, a cone with a flashing warning light is placed every 3 to 5 cones to enhance long-distance recognition.

[0038] The parameters of the spatial constraint model are calibrated by real-time traffic data. When the traffic flow suddenly increases to more than 70% of the road capacity, the length of the transition zone is automatically shortened by 10% to 15%, and the length of the warning zone is increased. When the proportion of large vehicles exceeds 20%, the lateral safety distance of the cones is increased by 0.3 to 0.5 meters. In rainy or snowy weather conditions, the spacing between cones is reduced and the number of cones with reflective materials is increased.

[0039] The location adaptation rules are coupled with the spatial constraint model to generate the target deployment location sequence. During the coupling calculation, the hard constraints of the spatial constraint model are satisfied first. Within the range allowed by the constraints, the location adaptation rules are applied. The generated target deployment location sequence includes the latitude and longitude coordinates, orientation angle and deployment priority of each cone.

[0040] Based on the current position of each cone unit, a bilateral matching iteration is used for position allocation. The bilateral matching iteration algorithm calculates the movement cost from each cone unit to each target deployment position. The movement cost takes into account both the movement distance and the movement time. During the iteration process, the nearest unallocated cone unit is allocated to the target position with the highest deployment priority in turn, until all target positions are allocated. The allocation result is sent to the corresponding cone unit through the wireless communication module. The allocation result includes the target deployment position coordinates and the arrival time window.

[0041] In practical applications, such as a highway bridge maintenance project requiring the closure of the right lane for three days, this method identifies the section as a two-way six-lane downhill section with a shoulder width of 2.5 meters and a longitudinal slope of 2.5%, averaging 90 km / h. Calculations determine the lateral safety distance of the mainline traffic cones to be 1.2 meters, the longitudinal spacing to be 20 meters, and the transition zone length to be 110 meters. During nighttime periods of reduced traffic flow, the intelligent traffic cone system automatically adjusts its deployment, increasing the longitudinal spacing to 25 meters and activating a flashing warning light combination mode, ensuring both traffic safety and construction efficiency during the work period.

[0042] In one optional implementation, the position adaptation rule is coupled with the spatial constraint model to generate a target deployment position sequence. Combined with the current position of each cone unit, a bilateral matching iteration is used for position allocation. The allocation result is then sent to the corresponding cone unit, including: The location adaptation rules are substituted into the constraints of the spatial constraint model for coupled calculation. Within the spatial range of the constraints, a set of candidate deployment locations is generated according to the location adaptation rules, and spatial continuity is checked. Transition location points are supplemented by interpolation to generate a sequence of target deployment locations. Obtain the current position coordinates and remaining power of each cone unit, calculate the travel time and energy consumption of each cone unit to reach each position point in the target deployment position sequence, generate a cone unit preference ranking table for each cone unit to the target deployment position sequence, and a target position preference ranking table for each target deployment position to the cone unit. Based on the bucket unit preference ranking table and the target location preference ranking table, a bilateral matching iteration is performed. Each bucket unit initiates a matching request to the target deployment position with the highest ranking in the bucket unit preference ranking table. Each target deployment position temporarily accepts the highest-ranked bucket unit and rejects other bucket units according to the target location preference ranking table. The rejected bucket unit initiates a new matching request to the second-highest ranked target deployment position. This iteration is repeated until all bucket units are accepted, and a location allocation result is generated and sent to the corresponding bucket unit.

[0043] In the process of automatic deployment of intelligent cones, the coupled calculation of position adaptation rules and spatial constraint models is a key step in determining the optimal deployment position. It is necessary to establish a road coordinate system with the center point of the construction area as the origin, the longitudinal direction of the road as the X-axis, and the lateral direction as the Y-axis to establish a two-dimensional coordinate system. Based on this coordinate system, the road feature data and deployment task parameters are digitized to form calculable spatial constraints.

[0044] During the coupled calculation process, it is necessary to determine the coordinates of the starting and ending points of the construction area, calculate the construction length, and determine the spatial range of the transition zone and buffer zone based on the lane closure method in the deployment task parameters. For example, in the case of closing the right lane, the Y coordinate of the construction area is located at the center line of the lane, and the X coordinate range is the start and end points of the construction. Under spatial constraints, the entire deployment area is rasterized, with the raster size set to 0.5 meters × 0.5 meters, to generate a set of candidate deployment location points.

[0045] The candidate placement locations are selected and optimized using location adaptation rules. In construction zones, cones are placed at the edges of closed lanes, with lateral spacing determined by lane width and safety distance. In transition zones, cones are placed in a linear transition, gradually shifting from the edge of unclosed lanes to the edge of closed lanes, with the shift angle typically controlled between 15° and 20°. In buffer zones, cones form a visual warning strip with a higher density than in other zones. Based on the aforementioned adaptation rules, points that meet the conditions are selected from the candidate placement locations to form a preliminary placement location sequence.

[0046] The initial placement sequence needs to undergo spatial continuity testing to ensure smooth distance changes between adjacent cones and avoid abrupt changes. The testing method is to calculate the distance difference between adjacent cones. If it exceeds a preset threshold (usually 20% of the baseline spacing), transition points are added using an interpolation algorithm. Cubic spline interpolation is used to ensure continuous path curvature and smooth visual guidance. For the connection between the starting point of the gradient area and the main line area, placement points are denser, and the spacing can be reduced to 70% of the baseline spacing to ensure a natural transition. After completing the spatial continuity test and interpolation supplementation, the final target placement sequence is formed, with each placement point containing precise coordinate information and cone orientation angle.

[0047] Obtaining real-time status information of each intelligent cone unit on site is a prerequisite for location allocation. This is achieved by receiving status data packets from each cone unit via a wireless communication network. These data packets contain information such as current location coordinates (longitude, latitude, and elevation), remaining battery percentage, communication signal strength, and equipment operating status codes. Location coordinates are obtained through the cone's built-in GPS / BeiDou positioning module, with a positioning accuracy typically within 1 meter. Remaining battery power is monitored in real-time by the battery management system, with an accuracy of 1%. Equipment operating status codes include various status codes such as normal, low battery, and fault.

[0048] Based on the sequence of the current position of the cones and the target deployment positions, the movement cost of each cone unit to reach each target position is calculated. The movement cost includes two parts: movement time and energy consumption. The movement time is calculated based on the maximum movement speed of the cone unit (usually 0.5 m / s) and the path length. Considering the influence of road conditions, the road friction coefficient is introduced to correct for the time. The energy consumption is estimated based on the cone movement power model, taking into account the movement distance, road slope and cone weight. It is usually expressed as a percentage of the remaining power consumption. For example, moving 100 meters on a flat road, a typical smart cone consumes about 5% of its power.

[0049] For each cone-shaped unit, the comprehensive cost is calculated for all locations in the target deployment sequence based on the calculated travel time and energy consumption. The locations are then sorted from low to high cost to generate a target location preference ranking table for that cone-shaped unit. The comprehensive cost is calculated using a weighted summation method, with travel time having a weight of 0.6 and energy consumption having a weight of 0.4. For cone-shaped units with less than 30% remaining power, the energy consumption weight is increased to 0.7 and the travel time weight is decreased to 0.3, prioritizing energy conservation.

[0050] Similarly, for each location point in the target deployment location sequence, the cone units are sorted from low to high according to the comprehensive cost of each cone unit to reach that location, and a cone unit preference ranking table for that target location is generated. For key location points, such as the starting point of the transition zone and the end point of the construction zone, the cone battery status and communication quality are additionally considered when ranking preferences, and cone units with high remaining battery and stable communication are given higher ranking priority.

[0051] Based on the bucket unit preference ranking table and the target position preference ranking table, a bilateral matching iterative algorithm is used for position allocation. The specific iterative process is as follows: Each bucket unit initiates a matching request to the target position with the highest ranking in its preference ranking table; each target position receives multiple requests from bucket units, and according to the bucket unit preference ranking table of that position, temporarily accepts the highest-ranked bucket unit and rejects the remaining requests; the rejected bucket unit moves to the next target position in its preference ranking table and initiates a new matching request; if the target position receives a new request with a higher ranking than the currently temporarily accepted bucket unit, it rejects the original bucket unit, accepts the new requester, and the rejected bucket unit continues to apply to the next target position; the above steps are repeated until all bucket units are accepted by a target position, or the preset maximum number of iterations is reached (usually 3 times the number of buckets).

[0052] After the bilateral matching iteration is completed, the matching result is the final location allocation scheme. This scheme has the characteristic of "stability", that is, there is no situation where the cone unit and the target position tend to match each other instead of the current match. For the final matching result, a location allocation instruction data packet is generated, which includes information such as cone ID, target position coordinates, and recommended movement path. The location allocation instruction is sent to the corresponding cone unit through the wireless communication network.

[0053] In practical applications, such as the deployment of intelligent traffic cones in highway construction sections, the entire calculation and allocation process can be completed within 5 seconds. For example, a highway needs to close the right lane for road repair, with a construction length of 200 meters, involving 30 intelligent traffic cone units. Based on this method, a target sequence containing 45 deployment locations is generated (15 in the construction area, 20 in the transition area, and 10 in the buffer zone); the movement cost matrix from the 30 cones to the 45 locations is calculated; through bilateral matching iteration, a stable match is achieved after 12 iterations, generating the final allocation scheme.

[0054] Figure 2 This is a flowchart illustrating the method for constructing a path conflict prediction matrix according to an embodiment of the present invention. In one optional implementation, each cone unit plans an initial movement trajectory based on its current position and the target deployment position. The initial movement trajectory is then transmitted to a collaborative planning node via a wireless communication module for spatiotemporal conflict detection. This process identifies overlapping trajectory segments and their collision times, and constructs a path conflict prediction matrix, including: An initial movement trajectory is planned based on the current position and the target deployment position. A trajectory envelope is constructed by combining the physical radius of the cone. The trajectory envelope is the boundary of a buffer zone with the initial movement trajectory as the center line and the physical radius of the cone as the offset distance. The trajectory envelope and the moving speed are sent to the collaborative planning node through the wireless communication module. The spatial intersection of the trajectory envelopes of each cone unit is determined pairwise. The overlapping area of ​​the paired trajectory envelopes is calculated, and the boundary coordinates of the overlapping area are extracted as the intersecting trajectory segments. Calculate the entry time of each cone unit into the overlapping area and the exit time from the overlapping area, determine whether the entry time and exit time of the paired cone units overlap, and when the entry time of one cone unit is earlier than the exit time of another cone unit and later than the entry time of another cone unit, record the start time of the time overlap interval as the collision time. Construct a path conflict prediction matrix, wherein the row index and column index of the path conflict prediction matrix correspond to each cone cell, and the matrix elements store the intersection and overlap trajectory segments and collision times of paired cone cells.

[0055] In this specific embodiment, after receiving the target deployment location allocation result, the intelligent cone needs to plan an initial movement trajectory based on its current position and the target deployment location. The road environment is discretized into a 0.5m × 0.5m grid, and obstacle areas are marked as impassable grids. Taking the current position of the cone as the starting point and the target deployment location as the ending point, a heuristic function is designed as a weighted sum of Euclidean distance and Manhattan distance, with weights of 0.4 and 0.6, respectively. By iteratively expanding the search space, the path sequence with the minimum cost is found, forming the initial movement trajectory point set.

[0056] To smooth the trajectory, cubic spline interpolation is applied to the initial set of moving trajectory points to generate a continuous and smooth curve. The sampling interval is 0.2 meters, forming a dense set of trajectory points. Based on the physical characteristics of the cone, the trajectory envelope is constructed considering the cone's radius. The standard intelligent cone base diameter is approximately 0.4 meters, corresponding to a radius of 0.2 meters. Using the interpolated trajectory as the centerline, the boundary of the buffer region, i.e., the trajectory envelope, is generated by offsetting the cone's physical radius to both sides. This envelope is represented by a polygon, and the vertex density is dynamically adjusted according to the curvature. The vertex density increases in areas with high curvature to ensure that the envelope accurately describes the space occupied by the cone during its movement.

[0057] The planned trajectory envelope information is sent to the collaborative planning node via a wireless communication module. The sent data includes the cone ID, trajectory envelope vertex coordinate sequence, expected movement speed, start timestamp, and expected arrival time. The wireless communication adopts the IEEE 802.11n protocol, the data packet size is usually within 20KB, and the transmission delay is controlled below 100 milliseconds. The collaborative planning node can be an edge computing unit or a cloud server on site, responsible for aggregating the trajectory information of all cones and performing collision detection.

[0058] At the collaborative planning node, pairwise spatial intersection determination is performed on the collected cone-shaped unit trajectory envelopes. A scan-line algorithm and bounding box fast filtering technique are used to improve determination efficiency. The minimum bounding rectangle of each trajectory envelope is calculated, and trajectory pairs that cannot intersect are quickly eliminated through rectangle intersection judgment. For intersecting trajectory pairs, a polygon intersection algorithm from planar computational geometry is applied to accurately determine whether the trajectory envelopes overlap. The time complexity of the intersection determination is O(n log n). 2 ), where n is the number of cones. For a scenario with 30 cones, the calculation time is usually within 200 milliseconds.

[0059] When it is determined that the envelopes of two trajectories intersect in space, the overlapping region is further extracted. The overlapping region is obtained by calculating the intersection of two polygons, and the result is also represented by a polygon. The boundary coordinates of the overlapping region, i.e., the sequence of polygon vertices, are extracted as the identifier of the intersecting trajectory segments. At the same time, the geometric center coordinates, area size and shape features of the overlapping region are calculated for subsequent collision risk assessment. For overlapping regions with complex shapes, polygon simplification is required to retain key feature points and reduce the amount of subsequent calculations.

[0060] Based on the cone's moving speed and trajectory length, the entry time of each cone unit into the overlapping area and the departure time from the overlapping area are calculated. The entry time is calculated by dividing the trajectory length from the cone's current position to the entrance point of the overlapping area by the cone's moving speed; the departure time is calculated by dividing the trajectory length from the cone's current position to the exit point of the overlapping area by the cone's moving speed. The typical moving speed of the intelligent cone is 0.5 m / s, and some high-performance cones can reach 0.8 m / s. To improve calculation accuracy, considering the cone's acceleration characteristics, the acceleration during the initial stage is approximately 0.2 m / s². 2 The deceleration during the deceleration phase is approximately 0.3 m / s². 2 .

[0061] For each pair of colliding cone units, their time overlap is determined. The criteria are: if the entry time of one cone unit is earlier than the exit time of another cone unit and later than the entry time of another cone unit, or if the entry times of the two cone units are the same, then there is a time overlap. In actual calculations, a time tolerance threshold is introduced, usually set to 0.5 seconds, to cope with positioning errors and speed fluctuations. When there is a time overlap, the start time of the overlap interval is recorded as the collision time, and the duration of the time overlap is calculated.

[0062] Based on the spatial and temporal overlap determination results, a path conflict prediction matrix is ​​constructed. This matrix is ​​an n×n square matrix, where n is the number of cones. The row and column indices correspond to the IDs of each cone unit. The elements of the matrix store the coordinates of the intersecting overlapping trajectory segments and the collision time of paired cone units. If there is no collision risk between paired cones, the corresponding matrix element is empty; if there is a collision risk, the coordinates of the overlapping area, the collision time, and the risk level are recorded. The risk level is comprehensively assessed based on the area of ​​the overlapping area and the duration of the temporal overlap, and is divided into three levels: low, medium, and high.

[0063] Taking a highway construction site as an example, five smart cones are simultaneously moved from the rest area to the construction area for deployment. The path conflict prediction matrix shows that the trajectories of cone #2 and cone #4 overlap in the area of ​​approximately 0.3 square meters near the coordinates (120.5, 35.2). Cone #2 is expected to enter this area 35.6 seconds after it starts moving, and cone #4 is expected to enter this area 37.2 seconds after it starts moving. The time overlap between the two lasts for 2.5 seconds, and the risk level is "medium".

[0064] Based on the path conflict prediction matrix, a collision avoidance strategy is implemented. For each pair of potential collision cones detected, an appropriate collision avoidance scheme is selected according to the collision risk level and cone priority. The collision avoidance schemes include two categories: time adjustment and path adjustment. Time adjustment is achieved by delaying the departure time of a cone or adjusting its movement speed to stagger the collision time window. For example, when there is a significant difference in cone priorities, time adjustment is preferred, delaying the departure time of the low-priority cone by at least 1.5 times the collision duration to ensure that the high-priority cone passes through the conflict area first.

[0065] Path adjustment involves modifying the cone's trajectory to bypass conflict areas. An elastic potential field method is used, treating the conflict area as a repulsive field and the original trajectory as an attractive field, calculating a new trajectory through force field equilibrium. The modification range is limited by the width of the feasible region, typically with an offset not exceeding 30% of the original trajectory. For complex scenarios with multiple conflict points, a global optimization strategy is employed, comprehensively considering all conflict points and adjusting the trajectory all at once to avoid local adjustments triggering new conflicts.

[0066] The collision avoidance scheme is applied to the initial trajectory to generate the final movement execution trajectory. Trajectory adjustment instructions, including the adjusted trajectory point set, velocity curve, and time control points, are sent to each cone unit. Upon receiving the instructions, the cone unit activates its built-in navigation control system and moves precisely along the planned trajectory. During movement, the cone continuously monitors its position; if it deviates from the planned trajectory by more than a threshold (typically 0.15m), it automatically adjusts its direction of movement to return to the planned trajectory. Simultaneously, the cone reports its position information to the collaborative planning node every 0.5 seconds for real-time trajectory monitoring and collision risk reassessment.

[0067] When an unforeseen obstacle or excessive deviation in trajectory execution is detected, an emergency collision avoidance mechanism is triggered. The cone immediately slows down to 30% of its moving speed or comes to a complete stop, and an abnormal status is reported. After receiving the abnormal status report, the collaborative planning node re-executes the trajectory planning and collision detection process, generates an emergency adjustment plan, and sends an alarm to the human operator for manual intervention in complex situations that cannot be resolved automatically. Through this multi-layered collision avoidance mechanism, the safety and reliability of the smart cone during automatic deployment are ensured.

[0068] In one optional implementation, based on the path conflict prediction matrix, the movement trajectories of conflicting cone units are replanned using a cooperative trajectory optimization algorithm. Different intermediate path points are assigned to the conflicting cone units, or their respective movement speeds are adjusted, to generate a conflict-free cooperative movement scheme, including: Extract the conflicting cone cells, overlapping trajectory segments, and collision times from the path conflict prediction matrix; A virtual gravitational field is constructed using the target deployment position of each cone unit as the gravitational source. The virtual gravitational field generates a gravitational vector on the cone unit pointing towards the target deployment position. A virtual repulsive field is constructed using the overlapping trajectory segments of conflicting cone units as the repulsive force source. The virtual repulsive field generates a repulsive force vector on other cone units that deviates from the overlapping trajectory segments. The virtual gravitational field and the virtual repulsive field are superimposed and synthesized to calculate the resultant force vector of each cone unit at the current position. The moving direction and moving speed of each cone unit are adjusted according to the resultant force vector. A bypass intermediate path point is generated at a position perpendicular to the repulsive force vector. The distance from the bypass intermediate path point to the boundary of the intersecting and overlapping trajectory segment is greater than the safety distance threshold. The updated movement trajectory of the cone unit is an adjusted trajectory that sequentially passes through the current position, detours around the intermediate path point, and the target deployment position; The potential field calculation and motion parameter update are performed iteratively. In each iteration, the resultant force vector, moving speed and bypass intermediate path point are recalculated until all cone units reach the target deployment position, generating a conflict-free cooperative movement scheme.

[0069] In this specific embodiment, after constructing the path conflict prediction matrix, it is necessary to extract information on conflicting cone units to effectively avoid collisions. This is done by traversing the path conflict prediction matrix and identifying the row and column indices corresponding to non-empty elements, which represent the conflicting cone unit pairs. Each non-empty matrix element contains a set of coordinates of intersecting and overlapping trajectory segments and collision time information. For each pair of conflicting cones, the coordinate sequence of the boundary points of their intersecting and overlapping trajectory segments is extracted, typically represented by polygons or sets of line segments. The collision time is also extracted, including the expected collision start time, duration, and end time. The extracted conflict information is organized into a conflict description list, with each list item containing a conflicting cone ID pair, a geometric description of the overlapping trajectory segments, and time window parameters, providing a data foundation for the subsequent formulation of collision avoidance strategies.

[0070] Based on the conflict description list, a virtual gravitational field is constructed using the artificial potential field method. The construction of the gravitational field uses the target deployment position of each cone unit as the gravitational source, and the magnitude of the gravity is inversely proportional to the distance from the cone to the target position. Specifically, the direction of the gravity vector F_attr points to the target deployment position, and its magnitude can be expressed as a function of the distance from the current position of the cone to the target position, usually calculated using a quadratic relationship. Under the action of the gravitational field, the cone unit naturally tends to move to the target position along the shortest path. The gravitational field parameters need to be adjusted according to the performance characteristics of the cone to ensure that the gravity does not cause the cone to move at excessive speed. The typical gravitational field strength coefficient is set in the range of [0.2, 0.5].

[0071] A virtual repulsive field is constructed synchronously to avoid potential collisions between cones. The repulsive field uses the overlapping trajectory segments with conflict as repulsive sources to generate repulsive vectors away from the conflict area for the cone units. The magnitude of the repulsive force adopts an exponential decay model, with the repulsive force being greater the closer to the repulsive source. After the distance exceeds the safety threshold, the repulsive force rapidly decays to a negligible level. The direction of the repulsive force vector F_rep is the opposite direction from the current position of the cone to the nearest point of the overlapping trajectory segment. To handle the case of multiple overlapping trajectory segments, the repulsive vectors generated by each repulsive source are combined into a total repulsive force vector through vector addition. The intensity coefficient of the repulsive field is usually set to 1.5 to 2 times that of the gravitational field to ensure that the repulsive force dominates when approaching the conflict area, effectively preventing the risk of collision.

[0072] The virtual gravitational field and the virtual repulsive field are superimposed and synthesized to calculate the resultant force vector of each cone unit at its current position. The resultant force vector F_total is the vector sum of the gravitational vector F_attr and the repulsive vector F_rep. The direction of the resultant force vector determines the direction of movement of the cone unit, and the magnitude of the resultant force determines the speed of movement of the cone unit. The speed of movement is proportional to the magnitude of the resultant force, but does not exceed the maximum speed limit of the cone (usually 0.5 m / s to 0.8 m / s). In the synthesized potential field, when the cone approaches the conflict area, the repulsive force increases, causing the direction of the resultant force to deviate from the original path, naturally forming a detour behavior; when it moves away from the conflict area, the gravity becomes dominant, causing the cone to point back to the target position. For complex scenarios with multiple conflict points, local minima traps may occur, which are resolved by introducing random perturbations or virtual obstacles to ensure that the cone can successfully reach the target position.

[0073] Based on the direction of the resultant force vector, a detour intermediate path point is generated in the direction perpendicular to the repulsive force vector to achieve smooth obstacle avoidance. The specific generation method is to calculate the unit vector of the repulsive force vector and rotate it 90 degrees to obtain the vertical direction. Along this vertical direction, with the boundary of the intersecting and overlapping trajectory segment as the reference point, a detour intermediate path point is generated by offsetting a certain distance. This distance should be greater than the preset safety distance threshold, which is usually 1.5 times the diameter of the cone (about 0.6 meters). In order to avoid the generated detour path point from falling into other conflict areas, the distance between the point and all conflict areas needs to be checked. If it is less than the safety threshold, the position is further adjusted. The selection of the detour direction is based on the relative relationship of the target position and the current road environment, giving priority to the side that is closer to the target position to reduce the path length increased by the detour.

[0074] The movement trajectory of the cone unit is updated, generating an adjusted path. This adjusted trajectory consists of three segments: a path segment from the current position to the intermediate path point, a path segment from the intermediate path point to the target deployment position, and a smooth transition curve at the connection point. The first path segment uses a straight line from the current position to the intermediate path point; the second path segment uses a straight line from the intermediate path point to the target position. The connection point uses a Bézier curve for a smooth transition with continuous curvature, preventing abrupt changes in curvature during cone turns. The total length of the adjusted trajectory is typically 10% to 20% longer than the original straight path, but effectively avoids potential collisions. Simultaneously, the cone's motion parameters, including the movement speed curve and turning angle sequence, are updated based on the new trajectory to ensure the cone accurately executes the adjusted trajectory.

[0075] After path adjustment, the process enters the iterative execution phase, continuously optimizing the cooperative movement scheme. Each iteration includes four steps: updating the cone position information, recalculating the resultant force vector, updating movement parameters, and checking termination conditions. The iteration cycle is typically set to 0.5 to 1 second, determined based on the cone's movement speed and control precision. In each iteration, based on the latest cone position, the potential field distribution and resultant force vector are recalculated, and the position of the intermediate path point is dynamically adjusted to achieve real-time response to environmental changes. Simultaneously, the cone's movement speed and direction are updated, and control commands are sent to the corresponding cone unit. The iterative process continues until the termination condition is met: all cone units reach their respective target deployment positions, or the preset maximum number of iterations is reached (usually the total path length divided by twice the distance moved per step).

[0076] In practical applications, such as the deployment of intelligent traffic cones in highway construction areas, the virtual potential field method demonstrates excellent collaborative obstacle avoidance performance. For example, a highway requires construction in the middle lane of a three-lane road, requiring 15 intelligent traffic cones to be moved from the roadside storage area to different deployment positions. Path conflict prediction identified three potential collision points: cones 1 and 5 intersect at the entrance of the transition zone 12 meters from the starting point; cones 6 and 9 face a risk of same-direction tracking in the middle of the construction zone 25 meters from the starting point; and cones 11 and 14 have a conflicting passage at the exit of the transition zone 40 meters from the starting point. After applying the virtual potential field method, cone 1 generates a detour path point offset 0.8 meters to the right, allowing cone 5 to pass first; cone 6 slows down by 20%, increasing the time interval with cone 9; and cone 11 delays its start by 2 seconds, avoiding the time window with cone 14. Ultimately, all cones safely reach their designated positions within 120 seconds without any collisions.

[0077] The complete cooperative movement scheme includes a sequence of trajectory points, a speed control curve, and time markers for each cone. The sequence of trajectory points is represented by coordinate points spaced 0.5 meters apart; the speed control curve describes the cone's running speed on each trajectory segment, taking into account acceleration and deceleration characteristics; the time markers record the estimated time when the cone reaches key points (such as detour points, turning points, and target positions). The cooperative movement scheme is distributed to each cone unit via a wireless network, and the cone controls its motor to operate according to the scheme to achieve precise positioning.

[0078] In one optional implementation, each cone unit moves according to the cooperative movement scheme and reports its location information, receives deployment change requests from the remote control terminal in real time, parses the target area parameters in the deployment change request, generates a new cone deployment position sequence, and issues movement commands, including: Each cone unit collects operating status parameters during movement and sends them to a remote control terminal. The operating status parameters include the current position coordinates, remaining power percentage, and remaining distance to the target deployment location. The remote control terminal calculates the energy adequacy index of each cone unit based on the remaining power percentage and remaining distance. The remote control terminal receives the deployment change request and extracts the new target area boundary coordinates, generates a new cone deployment position sequence, and calculates the estimated moving energy consumption of each cone unit from its current position coordinates to each candidate position in the new cone deployment position sequence. Construct a task allocation cost matrix, wherein the row index of the task allocation cost matrix is ​​the cone cell identifier, the column index is the position number in the new cone deployment position sequence, and the matrix element value is the weighted sum of the estimated mobile energy consumption value and the energy adequacy index; Based on the task allocation cost matrix, perform optimal matching calculation, allocate new target placement positions that minimize the global cost to each cone unit, and generate a redistribution instruction containing the cone unit identifier and the coordinates of the new target placement positions; The redistribution command is issued to each cone unit, triggering each cone unit to update the target deployment position and re-execute path planning.

[0079] In this specific embodiment, as each cone unit moves towards the target deployment location according to the cooperative movement scheme, its built-in sensor module continuously collects operational status parameters. The current position coordinates are obtained in real-time via a global satellite navigation system receiver, with a positioning accuracy better than 0.5 meters. The remaining battery percentage is calculated by the battery management chip after reading the voltage and current data of the lithium battery pack, with a sampling frequency of once per second. The remaining distance to the target deployment location is obtained by calculating the Euclidean distance between the current position coordinates and the target deployment location coordinates. Each cone unit encapsulates the above operational status parameters into data frames and transmits them to the remote control terminal via a 4G or 5G wireless communication module at a frequency of once every 5 seconds.

[0080] After receiving the operating status parameters, the remote control terminal calculates the energy adequacy index for each cone unit. This index is calculated by comparing the remaining power percentage with the theoretical power percentage required to complete the remaining distance; the difference is the energy adequacy index. The theoretical power percentage is determined by multiplying the cone unit's energy consumption per unit distance by the remaining distance. The energy consumption per unit distance is obtained through historical operating data statistics, with a typical value of 0.08% power consumption per meter. A positive energy adequacy index indicates that the cone unit has sufficient power; a negative value indicates that the unit cannot complete the current task.

[0081] When the road closure area needs to be adjusted, the remote control terminal's interface receives a deployment change request from the administrator. This request includes the new target area boundary coordinates, typically given as a sequence of polygon vertex coordinates. The remote control terminal parses these boundary coordinates and, based on parameters such as road width and cone spacing standards, generates a new sequence of cone placement positions along the boundary. The spacing setting is usually 3 to 5 meters, with the specific value determined according to the road grade.

[0082] For each candidate position in the newly generated cone deployment position sequence, the remote control terminal calculates the estimated movement energy consumption of each cone unit from its current position to that candidate position. The calculation first calculates the straight-line distance between the two points, then multiplies it by the energy consumption parameter per unit distance, and adds the additional energy consumption correction amount generated by turning and acceleration / deceleration. The correction amount is estimated based on the path curvature and speed change, with a typical correction coefficient of 1.2.

[0083] The remote control terminal constructs a task allocation cost matrix. The number of rows in this matrix equals the number of cone units currently participating in deployment, and the number of columns equals the number of positions in the new cone deployment position sequence. The value of the element in the i-th row and j-th column of the matrix is ​​calculated as follows: the estimated mobile energy consumption multiplied by a weighting coefficient of 0.6, and the negative of the energy adequacy index multiplied by a weighting coefficient of 0.4. The sum of these two values ​​yields the comprehensive cost. The weighting coefficients are set so that the system prioritizes the shortest travel distance while also considering the power balance of each cone unit.

[0084] Based on the constructed task allocation cost matrix, the remote control terminal invokes the Hungarian algorithm or auction algorithm to perform optimal matching calculations. The algorithm outputs a one-to-one matching scheme, ensuring that each cone cell is assigned to a unique new target deployment location, and minimizing the total global cost. After the calculation is complete, the system generates a reallocation instruction, which is encapsulated in JSON format and contains the unique identifier of the cone cell and the latitude and longitude coordinates of the new target deployment location.

[0085] The remote control terminal sends redistribution commands to each cone unit via a wireless communication network. Upon receiving the command, each cone unit immediately updates its locally stored target deployment location data, terminates its current movement task, and triggers the path planning module to recalculate the optimal movement trajectory from its current location to the new target deployment location. The replanned trajectory also needs to be uploaded to the collaborative planning node for conflict detection and collaborative optimization to ensure that the deployment process remains safe and orderly after the change.

[0086] The intelligent cone-shaped automatic deployment and remote collaborative control system of this invention includes: The information acquisition unit is used to acquire road feature information and deployment task parameters of the area to be defended. The location calculation unit is used to establish location adaptation rules and construct a spatial constraint model based on the road feature information and deployment task parameters, calculate the target deployment position of each smart cone and assign it to the corresponding cone unit. The conflict detection unit is used by each cone unit to plan an initial movement trajectory based on its current position and the target deployment position, and to send the initial movement trajectory to the collaborative planning node through the wireless communication module for spatiotemporal conflict detection, identify the trajectory segments that intersect and overlap and the collision time, and construct a path conflict prediction matrix. The trajectory optimization unit is used to replan the movement trajectory of conflicting cone units according to the path conflict prediction matrix and through a cooperative trajectory optimization algorithm, assign different intermediate path points to the conflicting cone units or adjust their respective movement speeds to generate a conflict-free cooperative movement scheme. The collaborative execution unit is used for each cone unit to move according to the collaborative movement scheme and report its position information, receive deployment change requests sent by the remote control terminal in real time, parse the target area parameters in the deployment change request, generate a new cone deployment position sequence and issue movement instructions.

[0087] A third aspect of the present invention provides an electronic device, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0088] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0089] This invention can be a method, apparatus, system, and / or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for performing various aspects of the invention.

[0090] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for automatic deployment and remote collaborative control of intelligent cone-shaped defenses, characterized in that, include: Obtain road feature information and deployment task parameters for the area to be defended; Based on the road feature information and deployment task parameters, location adaptation rules are established and a spatial constraint model is constructed. The target deployment positions of each smart cone are calculated and assigned to the corresponding cone units. Each cone unit plans an initial movement trajectory based on its current position and the target deployment position. The initial movement trajectory is then sent to the collaborative planning node via a wireless communication module for spatiotemporal conflict detection. This process identifies overlapping trajectory segments and collision times, and constructs a path conflict prediction matrix. Based on the path conflict prediction matrix, the movement trajectories of conflicting cone units are replanned using a cooperative trajectory optimization algorithm. Different intermediate path points are assigned to the conflicting cone units or their respective movement speeds are adjusted to generate a conflict-free cooperative movement scheme. Each cone unit moves according to the cooperative movement scheme and reports its location information. It receives deployment change requests from the remote control terminal in real time, parses the target area parameters in the deployment change request, generates a new cone deployment position sequence, and issues movement commands.

2. The method according to claim 1, characterized in that, Based on the road feature information and deployment task parameters, location adaptation rules are established and a spatial constraint model is constructed. The target deployment positions of each smart cone are calculated and assigned to the corresponding cone units, including: The number of lanes, shoulder width, longitudinal profile data and real-time traffic data are extracted from the road feature information. The cone placement level and lateral safety distance of the cones are determined according to the number of lanes and shoulder width. The longitudinal spacing of the cones is adjusted according to the real-time traffic data to obtain the position adaptation rules. The construction impact range and traffic organization plan are determined based on the deployment task parameters. The traffic organization plan includes lane closure methods and diversion paths. Combined with the longitudinal profile alignment data, a spatial constraint model for dynamic traffic flow guidance is constructed. The spatial constraint model includes cone gradient alignment constraints based on diversion paths, lateral position offset constraints based on lane closure methods, and cone sight distance guarantee constraints based on longitudinal profile slope. The constraint parameters of the spatial constraint model are calibrated using the real-time traffic data. The location adaptation rules are coupled with the spatial constraint model to generate a target deployment location sequence. Based on the current position of each cone unit, a bilateral matching iteration is used to allocate the location, and the allocation result is sent to the corresponding cone unit.

3. The method according to claim 2, characterized in that, The location adaptation rules are coupled with the spatial constraint model to generate a target deployment location sequence. Based on the current position of each cone unit, a bilateral matching iteration is used for location allocation. The allocation results are then sent to the corresponding cone unit, including: The location adaptation rules are substituted into the constraints of the spatial constraint model for coupled calculation. Within the spatial range of the constraints, a set of candidate deployment locations is generated according to the location adaptation rules, and spatial continuity is checked. Transition location points are supplemented by interpolation to generate a sequence of target deployment locations. Obtain the current position coordinates and remaining power of each cone unit, calculate the travel time and energy consumption of each cone unit to reach each position point in the target deployment position sequence, generate a cone unit preference ranking table for each cone unit to the target deployment position sequence, and a target position preference ranking table for each target deployment position to the cone unit. Based on the bucket unit preference ranking table and the target location preference ranking table, a bilateral matching iteration is performed. Each bucket unit initiates a matching request to the target deployment position with the highest ranking in the bucket unit preference ranking table. Each target deployment position temporarily accepts the highest-ranked bucket unit and rejects other bucket units according to the target location preference ranking table. The rejected bucket unit initiates a new matching request to the second-highest ranked target deployment position. This iteration is repeated until all bucket units are accepted, and a location allocation result is generated and sent to the corresponding bucket unit.

4. The method according to claim 1, characterized in that, Each cone unit plans an initial movement trajectory based on its current position and the target deployment position. This initial movement trajectory is then transmitted to the collaborative planning node via a wireless communication module for spatiotemporal conflict detection. This process identifies overlapping trajectory segments and their collision times, and constructs a path conflict prediction matrix, including: An initial movement trajectory is planned based on the current position and the target deployment position. A trajectory envelope is constructed by combining the physical radius of the cone. The trajectory envelope is the boundary of a buffer zone with the initial movement trajectory as the center line and the physical radius of the cone as the offset distance. The trajectory envelope and the moving speed are sent to the collaborative planning node through the wireless communication module. The spatial intersection of the trajectory envelopes of each cone unit is determined pairwise. The overlapping area of ​​the paired trajectory envelopes is calculated, and the boundary coordinates of the overlapping area are extracted as the intersecting trajectory segments. Calculate the entry time of each cone unit entering the overlapping region and the exit time of each cone unit leaving the overlapping region. Determine whether there is a time overlap between the entry and exit times of paired cone units. When the entry time of one cone unit is earlier than the exit time of another cone unit but later than the entry time of another cone unit, record the start time of the time overlap interval as the collision time. Construct a path conflict prediction matrix. The row index and column index of the path conflict prediction matrix correspond to each cone unit. The matrix elements store the intersection and overlap trajectory segments and collision times of paired cone units.

5. The method according to claim 1, characterized in that, Based on the path conflict prediction matrix, the movement trajectories of conflicting cone units are replanned using a cooperative trajectory optimization algorithm. Different intermediate path points are assigned to the conflicting cone units, or their respective movement speeds are adjusted, generating a conflict-free cooperative movement scheme, including: Extract the conflicting cone units, overlapping trajectory segments, and collision times from the path conflict prediction matrix; construct a virtual gravitational field using the target deployment position of each cone unit as the gravitational source, and the virtual gravitational field generates a gravitational vector pointing to the target deployment position on the cone units; A virtual repulsive field is constructed using the overlapping trajectory segments of conflicting cone units as the repulsive force source. The virtual repulsive field generates a repulsive force vector on other cone units that deviates from the overlapping trajectory segments. The virtual gravitational field and the virtual repulsive field are superimposed and synthesized to calculate the resultant force vector of each cone unit at the current position. The moving direction and moving speed of each cone unit are adjusted according to the resultant force vector. A bypass intermediate path point is generated at a position perpendicular to the repulsive force vector. The distance from the bypass intermediate path point to the boundary of the intersecting and overlapping trajectory segment is greater than the safety distance threshold. The movement trajectory of the cone unit is updated to an adjusted trajectory that passes through the current position, bypasses the intermediate path point, and the target deployment position in sequence; the potential field calculation and motion parameter update are performed iteratively, and the resultant force vector, movement speed, and bypass intermediate path point are recalculated in each iteration until all cone units reach the target deployment position, generating a conflict-free cooperative movement scheme.

6. The method according to claim 1, characterized in that, Each cone unit moves according to the aforementioned coordinated movement scheme and reports its location information. It receives deployment change requests from the remote control terminal in real time, parses the target area parameters in the deployment change request, generates a new cone deployment position sequence, and issues movement commands, including: During the movement of each cone unit, operating status parameters are collected and sent to the remote control terminal. The operating status parameters include the current position coordinates, remaining power percentage, and remaining distance to the target deployment position. The remote control terminal calculates the energy adequacy index of each cone unit based on the remaining power percentage and remaining distance. The remote control terminal receives a deployment change request and extracts the new target area boundary coordinates, generates a new cone deployment position sequence, calculates the estimated mobile energy consumption from the current position coordinates of each cone unit to each candidate position in the new cone deployment position sequence, and constructs a task allocation cost matrix. The row index of the task allocation cost matrix is ​​the cone unit identifier, the column index is the position number in the new cone deployment position sequence, and the matrix element value is the weighted sum of the estimated mobile energy consumption and the energy adequacy index. Based on the task allocation cost matrix, perform optimal matching calculation, allocate a new target placement position that minimizes the global cost to each cone unit, generate a reassignment instruction containing the cone unit identifier and the coordinates of the new target placement position, and issue the reassignment instruction to each cone unit to trigger each cone unit to update the target placement position and re-execute path planning.

7. An intelligent cone-shaped automatic deployment and remote collaborative control system, used to implement the method as described in any one of claims 1-6, characterized in that, include: The information acquisition unit is used to acquire road feature information and deployment task parameters of the area to be defended. The location calculation unit is used to establish location adaptation rules and construct a spatial constraint model based on the road feature information and deployment task parameters, calculate the target deployment position of each smart cone and assign it to the corresponding cone unit. The conflict detection unit is used by each cone unit to plan an initial movement trajectory based on its current position and the target deployment position, and to send the initial movement trajectory to the collaborative planning node through the wireless communication module for spatiotemporal conflict detection, identify the trajectory segments that intersect and overlap and the collision time, and construct a path conflict prediction matrix. The trajectory optimization unit is used to replan the movement trajectory of conflicting cone units according to the path conflict prediction matrix and through a cooperative trajectory optimization algorithm, assign different intermediate path points to the conflicting cone units or adjust their respective movement speeds to generate a conflict-free cooperative movement scheme. The collaborative execution unit is used for each cone unit to move according to the collaborative movement scheme and report its position information, receive deployment change requests sent by the remote control terminal in real time, parse the target area parameters in the deployment change request, generate a new cone deployment position sequence and issue movement instructions.

8. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 6.

Images