A movable travel and tourism night tour light show control device and control method

Abstract

This application provides a mobile control device and method for cultural tourism nighttime light shows. The method involves acquiring a state dataset of a group of light devices; determining a 3D point cloud digital model of the performance area; and, based on the 3D point cloud digital model and a preset performance timeline, determining the dynamic priority of each light and shadow performance task in the performance area. Each mobile light show device in the light device group is abstracted as a particle, and a decision population is constructed. Through the state dataset and the dynamic priority of each light and shadow performance task, the decision population is iteratively optimized using an adaptive search mode switching particle swarm optimization to obtain a collaborative control strategy for all mobile light show devices to perform light shows. Based on the collaborative control strategy, each mobile light show device is driven to perform an adaptive and collaborative group light show. Using the scheme of this application, a collaborative control method for mobile light shows integrating real-time state perception, dynamic task scheduling, and swarm intelligence optimization can be realized.

Description

Technical Field

[0001] This application relates to the field of swarm intelligence algorithm application technology, and more specifically, to a mobile control device and control method for cultural tourism nighttime light shows. Background Technology

[0002] With the rapid development of the cultural tourism industry, the nighttime economy has become an important growth point for consumption. As the core carrier of nighttime cultural tourism projects, light shows are evolving from traditional static and fixed sets to dynamic, interactive, and mobile ones. Mobile light show installations can break spatial limitations and present more immersive and narrative light and shadow performances through flexible formations and movements, bringing tourists a novel viewing experience. Against this backdrop, how to achieve efficient, adaptive, and collaborative control of multiple mobile light installations in complex performance environments has become the key to improving the artistic expression and technical reliability of light shows.

[0003] Currently, most existing mobile light shows employ pre-programmed path control or localized collaborative methods based on simple rules (such as following and obstacle avoidance). These methods lack global optimization and dynamic adjustment capabilities for inter-device collaboration. They typically rely on fixed performance scripts and centralized control, making it difficult to adapt to real-time environmental changes (such as audience distribution and device status) and dynamic adjustments to the performance plot. This results in low collaborative efficiency, poor fault tolerance, and an inability to intelligently allocate resources based on task priorities. Furthermore, traditional methods lack multi-objective optimization capabilities under multiple constraints (such as device energy, motion smoothness, and performance sequence), often failing to generate safe, energy-efficient, and artistically expressive group collaborative strategies within a limited timeframe. Therefore, achieving a collaborative control method for mobile light shows that integrates real-time status perception, dynamic task scheduling, and swarm intelligence optimization has become a significant challenge for the industry. Summary of the Invention

[0004] This application provides a mobile cultural tourism nighttime light show control device and control method, which can realize a mobile light show collaborative control method that integrates real-time status perception, dynamic task scheduling and swarm intelligence optimization.

[0005] Firstly, this application provides a control method for a mobile light show based on dynamic group collaboration, used to control a mobile cultural tourism nighttime light show device to perform a dynamic light and shadow narrative performance in a group collaboration manner. The method includes: The pose and energy state information of each movable light show device in the light installation group are obtained, and then the state dataset of the light installation group is obtained. A point cloud 3D model is performed on the target cultural tourism night tour area to obtain a 3D point cloud digital model of the performance area. Then, based on the 3D point cloud digital model and the preset performance plot timeline, the dynamic priority of each light and shadow performance task in the performance area is determined. Each mobile light show device is abstracted as a particle, and a decision population is constructed based on all particles. The decision population is then subjected to adaptive search mode switching particle swarm optimization iterative solution through the state dataset and the dynamic priority of each light and shadow performance task to obtain the collaborative control strategy for all mobile light show devices to perform light performances. The execution instruction set for each mobile light show device is determined according to the collaborative control strategy, and then each mobile light show device is driven to perform an adaptive and collaborative group light show within the performance area based on all the execution instruction sets.

[0006] In some embodiments, performing point cloud 3D modeling on the target cultural tourism night tour area to obtain a 3D point cloud digital model of the performance area specifically includes: Multi-view scanning of the target cultural tourism night tour area yielded the original three-dimensional point cloud data; A dense, colored point cloud covering the entire performance area is generated based on the original 3D point cloud data. Key environmental elements are identified in the dense colored point cloud, thereby constructing a three-dimensional point cloud digital model of the performance area.

[0007] In some embodiments, determining the dynamic priority of each light and shadow performance task in the performance area based on the three-dimensional point cloud digital model and a preset performance plot timeline specifically includes: The walkable region segmentation and visual focus extraction are performed on the three-dimensional point cloud digital model to obtain the key performance location point set of the performance area; Based on the preset performance plot timeline and the set of key performance locations, determine multiple light and shadow performance tasks associated with key performance locations; The urgency of each light and shadow performance task in terms of time is determined by identifying the key performance location points associated with all light and shadow performance tasks. Based on the urgency of all needs, tasks are sorted to obtain the dynamic priority of each light and shadow performance task in the performance area.

[0008] In some embodiments, abstracting each movable light show device as a particle, and then constructing a decision population based on all particles specifically includes: Each movable light show installation is abstracted as a particle; Based on the state dataset, determine the multi-dimensional state parameters of each particle; Using the multi-dimensional state parameters of each particle as initial constraints, a decision population is constructed based on the dynamic priority of all light and shadow performance tasks.

[0009] In some embodiments, the collaborative control strategy for all movable light show devices to perform light shows is obtained by iteratively solving the particle swarm optimization problem with adaptive search mode switching based on the state dataset and the dynamic priority of each light show task, specifically including: A multi-objective optimization function for collaborative control of lighting performances is constructed based on the state dataset and the dynamic priority of each lighting performance task. Particle swarm optimization iteratively is performed based on the decision population and the multi-objective optimization function; For each iteration, the evolutionary factor is determined based on the population distribution characteristics of the particle swarm; Based on the evolutionary factors, the search modes are dynamically switched to update the multi-dimensional state parameters of each particle, thereby obtaining the state parameter set of the particle swarm, and then obtaining the state parameter set of the particle swarm after each iteration. Once the iteration meets the termination condition, the iteration terminates and a collaborative control strategy for all movable light show devices to perform a light show is generated based on the state parameter set after the iteration ends.

[0010] In some embodiments, determining the set of execution instructions for each movable light show device according to the collaborative control strategy specifically includes: For each movable light show device, the motion control command of the movable light show device is determined according to the target position queue and motion speed curve of the movable light show device in the cooperative control strategy. The light drive command for the movable light show device is generated based on the sequence of light effect parameters of the movable light show device in the collaborative control strategy. A unified high-precision time synchronization signal is injected into the motion control command and the light driving command, and packaged to generate an execution command set for the mobile light show device, thereby obtaining the execution command set for each mobile light show device.

[0011] In some embodiments, driving each movable light show device to perform an adaptive and coordinated group light show within the performance area based on all execution instruction sets specifically includes: All execution instruction sets are distributed to the device controllers of each mobile light show device through a distributed synchronous execution architecture. The controllers of each movable light show device execute motion servoing and light flux output in parallel according to the received execution instruction set, so as to realize adaptive and coordinated group light show.

[0012] Secondly, this application provides a portable control device for cultural tourism nighttime light shows, the device comprising: The acquisition module is used to acquire the pose and energy state information of each movable light show device in the light installation group, thereby obtaining the state dataset of the light installation group. The perception and decision-making module is used to perform point cloud 3D modeling of the target cultural tourism night tour area to obtain a 3D point cloud digital model of the performance area, and then determine the dynamic priority of each light and shadow performance task in the performance area based on the 3D point cloud digital model and the preset performance plot timeline. The perception and decision-making module is used to abstract each mobile light show device as a particle, and then construct a decision population based on all particles. The decision population is then optimized iteratively by adaptively searching the state dataset and the dynamic priority of each light and shadow performance task to obtain the collaborative control strategy for all mobile light show devices to perform light performances. The execution module is used to determine the execution instruction set of each mobile light show device according to the collaborative control strategy, and then drive each mobile light show device to perform an adaptive and collaborative group light show within the performance area based on all the execution instruction sets.

[0013] Thirdly, this application provides a computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described mobile light show control method based on dynamic group collaboration.

[0014] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned mobile light show control method based on dynamic group collaboration.

[0015] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects: The movable cultural tourism nighttime light show control device and method provided in this application first acquires the pose and energy state information of each movable light show device in the light device group, thereby obtaining a state dataset of the light device group; performs point cloud 3D modeling on the target cultural tourism nighttime area to obtain a 3D point cloud digital model of the performance area, and then determines the dynamic priority of each light and shadow performance task in the performance area based on the 3D point cloud digital model and the preset performance plot timeline; abstracts each movable light show device as a particle, and then constructs a decision population based on all particles, and performs adaptive search mode switching particle swarm optimization iterative solution on the decision population through the state dataset and the dynamic priority of each light and shadow performance task to obtain a collaborative control strategy for all movable light show devices to perform light performances; determines the execution instruction set of each movable light show device according to the collaborative control strategy, and then drives each movable light show device to perform an adaptive and collaborative group light performance in the performance area based on all execution instruction sets.

[0016] Therefore, this application drives each movable light show device to perform an adaptive and coordinated group light show within the performance area based on all execution instruction sets. First, determining the state dataset yields a unified data set characterizing the instantaneous operating status of the entire group of light devices. This state dataset is determined by synchronously collecting, aggregating, and fusing the real-time pose (position and attitude) and energy status (such as remaining battery power) information of each movable light show device in the group, forming a unified, accurate, and spatiotemporally consistent system state snapshot. This lays a crucial data foundation for subsequent group collaborative decision-making, directly addressing the problems of rigid collaboration, delayed response, and poor fault tolerance caused by the lack of real-time and comprehensive device state awareness in existing technologies. This allows the collaborative control strategy to be dynamically generated and adjusted based on the actual physical state of the devices (such as position, movable range, and remaining battery power), significantly improving the overall adaptability, operational safety, and resource utilization efficiency of the light show system, achieving a fundamental shift from fixed programming to state-driven adaptive collaboration. Then, determining the dynamic priority yields a unified data set based on all light and shadow performance tasks. The system generates a globally ordered and real-time updated task execution order based on the urgency of the demand and the logical dependencies between tasks. The dynamic priority determination can be achieved by integrating the 3D point cloud digital model of the performance area, the preset plot timeline, and real-time perceived spatial attraction and audience distribution data. This transforms the originally static and fixed performance task sequence into a dynamic task queue that can evolve in real time with the environment and audience interaction. This mechanism directly addresses the problems of "poor adaptability of pre-programmed scripts and inability to respond to real-time changes" in existing technologies. By introducing dynamic sorting based on urgency quantification and task dependency analysis, the system can autonomously identify and prioritize the scheduling of the most critical and appropriate performance tasks at different times. This not only significantly improves the real-time responsiveness and narrative coherence of the light show performance, but also provides clear and time-varying optimization targets for downstream swarm intelligence optimization algorithms. This enables subsequent particle swarm optimization to efficiently generate collaborative strategies with reasonable resource allocation and coordinated actions under multiple constraints. As a result, the system achieves a leap from fixed process playback to environmental perception, dynamic decision-making, and collaborative execution, greatly enhancing the intelligence, adaptability, and overall performance efficiency of the system.Finally, determining the collaborative control strategy yields a complete set of instructions specifying the specific actions and light effects output of all mobile lighting devices in the spatiotemporal dimension. This collaborative control strategy can be determined by abstracting each mobile lighting device as a particle and iteratively solving it using a particle swarm optimization algorithm with adaptive search mode switching, based on real-time collected device status data and dynamic task priorities. This achieves global collaborative decision-making under multiple constraints. Therefore, the system can dynamically generate a group control scheme that balances motion safety, energy efficiency, and artistic expression based on the performance plot, real-time device status, and environmental changes. This significantly improves the collaborative intelligence level, adaptive adjustment capability, and overall performance effect of the light show system, effectively overcoming the problems of low flexibility, lag response, and single optimization objective inherent in traditional pre-programmed and simple rule-based collaboration. In summary, based on the above scheme, a collaborative control method for mobile light shows integrating real-time status perception, dynamic task scheduling, and swarm intelligence optimization can be realized. Attached Figure Description

[0017] Figure 1 This is an exemplary flowchart of a mobile light show control method based on dynamic group collaboration, as shown in some embodiments of this application. Figure 2 This is an exemplary flowchart illustrating the determination of dynamic priority according to some embodiments of this application; Figure 3 This is an operational flowchart illustrating the determination of a collaborative control strategy according to some embodiments of this application; Figure 4 This is a structural schematic diagram of a movable cultural tourism nighttime light show control device according to some embodiments of this application; Figure 5 This is an internal structural diagram of a computer device that implements a mobile light show control method based on dynamic group collaboration, according to some embodiments of this application. Detailed Implementation

[0018] To better understand the technical solution of this application, the technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0019] refer to Figure 1 The figure is an exemplary flowchart of a mobile light show control method based on dynamic group collaboration, according to some embodiments of this application. The mobile light show control method based on dynamic group collaboration mainly includes the following steps: In step 101, the pose and energy state information of each movable light show device in the light device group are obtained, thereby obtaining the state dataset of the light device group.

[0020] It should be noted that, in this application, the state dataset is a unified data set characterizing the instantaneous operating status of the entire group of lighting installations. It provides accurate, synchronous, and complete system state input for subsequent collaborative decision-making and control, thereby ensuring the real-time performance, safety, and overall effectiveness of the collaborative scheme. Specifically, obtaining the pose and energy state information of each movable light show device in the group, and thus obtaining the state dataset of the group, can be achieved in the following way: First, the linear acceleration and angular velocity of each movable light show device can be measured by an inertial measurement unit integrated on each device, and the raw satellite observation data obtained from the global navigation satellite system receiver can be fused and calculated to obtain the three-dimensional space of each movable light show device in a unified world coordinate system. Position, three-axis attitude angles, etc., constitute precise pose state information; secondly, the voltage, current, and temperature of the battery can be sampled and monitored in real time by the battery management system connected to the power battery of each mobile light show device, and estimated based on the battery electrochemical model to obtain its remaining capacity and estimated remaining driving time as precise energy state information; then, through a low-latency, high-reliability vehicle-to-vehicle wireless communication network deployed between devices, the pose state information and energy state information reported by each mobile light show device with local timestamps are transmitted and collected across nodes. The central coordinating controller or designated master node performs timestamp alignment and data fusion processing on all collected asynchronous information streams, and the processed, spatiotemporally consistent set of all information is used as the state dataset of the light show device group.

[0021] In step 102, a point cloud 3D model is performed on the target cultural tourism night tour area to obtain a 3D point cloud digital model of the performance area. Then, based on the 3D point cloud digital model and the preset performance plot timeline, the dynamic priority of each light and shadow performance task in the performance area is determined.

[0022] In some embodiments, the following steps can be taken to perform point cloud 3D modeling on the target cultural tourism night tour area to obtain a 3D point cloud digital model of the performance area: Multi-view scanning of the target cultural tourism night tour area yielded the original three-dimensional point cloud data; A dense, colored point cloud covering the entire performance area is generated based on the original 3D point cloud data. Key environmental elements are identified in the dense colored point cloud, thereby constructing a three-dimensional point cloud digital model of the performance area.

[0023] In practice, the target cultural tourism night tour area is scanned from multiple perspectives to obtain the original 3D point cloud data. This can be achieved in the following way: First, a 3D LiDAR sensor array deployed on a ground-based mobile mapping robot can be used to perform a pre-planned, multi-site, multi-angle scan of the target cultural tourism night tour area without blind spots, obtaining an original discrete point cloud set containing geometric coordinates and reflection intensity information collected from different spatial perspectives. Second, point cloud registration technology (such as the iterative nearest point algorithm) is used to perform spatial alignment and coordinate unification processing on the original discrete point cloud set, forming a complete but non-existent point cloud data set. The initial fused point cloud is obtained by eliminating data overlap and noise. Then, noise is filtered from the initial fused point cloud using a statistical outlier filtering algorithm to remove floating points caused by measurement errors. A voxelized mesh downsampling method is applied to reduce data redundancy while maintaining geometric features, resulting in a complete point cloud set with a unified spatial reference system, which serves as the original 3D point cloud data required for subsequent environmental reconstruction. The original 3D point cloud refers to a discrete set of points containing the spatial geometric coordinates of the target cultural tourism night tour area, which can provide a high-precision, quantifiable basic spatial data source faithful to the physical world for the entire environmental modeling process.

[0024] In specific implementation, generating a dense colored point cloud covering the entire performance area based on the original 3D point cloud data can be achieved in the following way: First, a high-resolution camera simultaneously mounted during multi-view scanning can acquire a sequence of color images strictly synchronized with each frame of laser point cloud data in time. Using the pre-calibrated extrinsic and intrinsic parameters between the camera and the lidar, each colored pixel is mapped to the corresponding 3D laser point cloud coordinates through perspective projection, giving the geometric point cloud accurate color attributes and generating a sparse colored point cloud. Then, a surface reconstruction algorithm (such as a surface reconstruction algorithm based on Poisson reconstruction or moving least squares) is used to interpolate and smooth the sparse colored point cloud, especially to reasonably infer and fill in the geometric and color missing areas of the point cloud caused by occlusion, thereby generating a dense colored point cloud with a continuous surface appearance and rich color information. Here, the dense colored point cloud refers to a continuous surface model reconstructed by fusing real color texture information on the basis of the original point cloud. It can provide an accurate geometric framework and intuitive visual context for environmental perception and understanding, and is a key input for scene semantic parsing.

[0025] In specific implementation, the key environmental elements of the dense colored point cloud are identified, and a three-dimensional point cloud digital model of the performance area is constructed in the following way: First, a point cloud semantic segmentation model based on a deep convolutional neural network is used to classify each point in the dense colored point cloud point by point, automatically identifying and labeling key environmental element categories, including passable ground, fixed building facades, vegetation, water bodies, and fixed lighting devices. Based on the semantic segmentation results, discrete point clouds belonging to the same category are aggregated into independent object instances with clear boundaries. At the same time, the instance geometric information of each instance is calculated, including the three-dimensional bounding box, centroid, surface normal vector, etc. Then, based on the semantic category results extracted by the point cloud semantic segmentation model and the instance geometric information, an environment model is constructed in a hierarchical structure. The top layer is the overall scene, and the lower layers are the category layers and specific instance objects. Each specific instance object is associated with its semantic label, geometric parameters, and original point cloud data, resulting in a hierarchical three-dimensional digital model that integrates accurate geometry, realistic texture, rich semantics, and instantiated structural information as the three-dimensional point cloud digital model of the performance area.

[0026] It should be noted that, in this application, the three-dimensional point cloud digital model refers to a hierarchical digital representation of the environment with object categories and attribute labels. It can transform the physical scene into a digital workspace that can be recognized and reasoned about by machines, providing a directly operable high-dimensional information base for subsequent light and shadow task space mapping, collaborative path planning and collision detection.

[0027] In some embodiments, reference Figure 2 The figure is an exemplary flowchart illustrating the determination of dynamic priority according to some embodiments of this application. The determination of the dynamic priority of each light and shadow performance task in the performance area based on the three-dimensional point cloud digital model and a preset performance plot timeline can be achieved through the following steps: In step 1021, the walkable region is segmented and the visual focus is extracted from the three-dimensional point cloud digital model to obtain the key performance location point set of the performance area; In step 1022, multiple light and shadow performance tasks associated with key performance locations are determined based on the preset performance plot timeline and the set of key performance locations. In step 1023, the urgency of each light and shadow performance task in the time dimension is determined by the key performance location points associated with all light and shadow performance tasks; In step 1024, tasks are sorted based on the urgency of all requirements to obtain the dynamic priority of each light and shadow performance task in the performance area.

[0028] In specific implementation, the walkable area segmentation and visual focus extraction of the 3D point cloud digital model to obtain the key performance location point set of the performance area can be achieved in the following way: First, walkable ground point clouds and potential obstacle point clouds, including fixed building facades, vegetation, water bodies, and fixed lighting devices, can be separated from the 3D point cloud digital model using point cloud clustering and segmentation technology based on element identification, to obtain a continuous walkable area patch for safe navigation of the movable light show device; Second, candidate performance planes are screened within the continuous walkable area patch through point cloud normal analysis and local curvature calculation, that is, large areas with gentle ground curvature and normal direction pointing upwards (i.e., approximately perpendicular to the horizontal plane). The area is then used to score and rank all candidate performance planes based on preset global visual evaluation criteria (such as regional centrality, distance from the main viewing path, and area). Multiple candidate performance planes with scores higher than 9 are selected, and their geometric centers are projected onto a two-dimensional plane to generate a buffer. The set of coordinates of the center points within the buffer is used as the key performance location point set for the performance area. The key performance location point set is the specific coordinate space anchor point that anchors the abstract performance plot to the three-dimensional physical space. It can provide a clear set of spatial targets for collaborative planning, ensuring that all subsequent performance tasks and actions are based on geographical locations that are actually reachable and have the best visual effect.

[0029] In specific implementation, determining multiple related key performance location points for light and shadow performance tasks based on a preset performance plot timeline and the set of key performance location points can be achieved in the following way: First, the preset performance plot timeline script can be parsed and broken down into a series of plot segments with clear start and end times, expected emotional tone, and light and shadow expression themes. For each plot segment, one or more abstract task action prototypes (such as fixed-point coloring, path scanning, pattern projection) can be matched from a preset task type template library according to its light and shadow expression theme (e.g., focusing, atmosphere rendering, dynamic tracking). Specific lighting effect parameter constraints (including...) can be configured for each task action prototype based on the plot segment's time attributes and emotional intensity. (Color temperature range, brightness curve, and frequency of change); then, through a spatiotemporal binding algorithm, each configured task action prototype is assigned and bound to one or more designated locations in the key performance location set according to the spatial narrative intent of its plot segment, generating multiple light and shadow performance tasks associated with key performance locations; wherein, the light and shadow performance task is a light and shadow performance atomic execution unit with clear time windows, spatial anchor points, and effect parameter constraints, which can deconstruct a complete plot segment or artistic intent in the narrative script into basic lighting actions that can be independently executed by a single movable light show device at a designated time window and spatial anchor point, thereby transforming complex artistic narratives into standard technical instructions that can be optimized and allocated by swarm intelligence algorithms.

[0030] In practice, determining the urgency of each light and shadow performance task in terms of time dimension by using key performance location points associated with all light and shadow performance tasks can be achieved in the following way: First, by performing time-series weighted calculations based on an exponential decay function on historical interaction data (including task binding frequency and audience dwell time) for each key performance location point, a basic spatial attraction value reflecting its long-term popularity is obtained. Then, by generating a heatmap based on kernel density estimation from real-time collected audience location data, the audience density value for each key performance location point is obtained. Finally, for each light and shadow performance task, the basic spatial attraction values ​​of all key performance location points associated with the light and shadow performance task are obtained and... The audience density value is calculated by averaging all spatial attraction base values ​​and all audience density values. The two average values ​​are then linearly weighted and fused with the difference between the planned trigger time of the light and shadow performance task and the current system time based on a preset weight coefficient to obtain the urgency of the light and shadow performance task in the time dimension. Through the above steps, the urgency of the demand for each light and shadow performance task in the time dimension can be obtained. The urgency of the demand is a decision factor that quantifies the degree of urgency of the light and shadow performance task to be prioritized due to spatial attraction and audience gathering. It can introduce the system with the ability to respond to environmental changes and audience interaction in real time, so that the decision basis changes from static preset to dynamic perception.

[0031] It should be noted that in this application, the dynamic priority is a globally ordered and real-time updated task execution order generated based on the urgency of all lighting and shadow performance tasks and the logical dependencies between tasks. It provides a core and dynamically changing optimization objective for downstream group collaborative optimization algorithms and is a key input driving the entire system to achieve adaptive collaborative choreography. Specifically, the dynamic priority of each lighting and shadow performance task in the performance area can be achieved by sorting tasks based on their urgency, as follows: First, by parsing the preset performance plot timeline, the temporal dependencies and logical constraints between lighting and shadow performance tasks (e.g., the atmosphere rendering task must precede the focus reveal task) are formalized into a directed acyclic graph, and then all lighting and shadow performance tasks are processed... Based on the topological reachability analysis of the directed acyclic graph, the execution sequence obtained from the analysis is used as a rigid precondition that must be followed in the sorting process. Under the premise of strictly satisfying the rigid precondition, the urgency of all light and shadow performance tasks is sorted in descending order to obtain a preliminary task execution order list with urgency as the core indicator. Then, in order to enhance the stability of the system in continuous decision-making and avoid resource allocation jitter, a time sliding window mechanism can be introduced to perform weighted averaging and consistency verification on the preliminary task order generated by multiple consecutive calculation cycles within the window, and priority inertia smoothing is implemented to generate a globally ordered and dynamically evolving priority list that evolves smoothly over time. This gives the dynamic priority of each light and shadow performance task in the performance area.

[0032] In step 103, each mobile light show device is abstracted as a particle, and a decision population is constructed based on all the particles. The decision population is then subjected to adaptive search mode switching particle swarm optimization iterative solution through the state dataset and the dynamic priority of each light and shadow performance task to obtain the collaborative control strategy for all mobile light show devices to perform light performances.

[0033] In some embodiments, abstracting each movable light show device as a particle and then constructing a decision population based on all particles can be achieved through the following steps: Each movable light show installation is abstracted as a particle; Based on the state dataset, determine the multi-dimensional state parameters of each particle; Using the multi-dimensional state parameters of each particle as initial constraints, a decision population is constructed based on the dynamic priority of all light and shadow performance tasks.

[0034] In practical implementation, each mobile light show device can be abstracted as a particle in the following way: a mapping model from physical entities to algorithmic intelligent agents can be established at the algorithm level of the control system, and each mobile light show device can be uniquely associated with an abstract intelligent agent, i.e., a particle, in a swarm intelligence algorithm; wherein, the particle is an abstract agent and autonomous decision-making unit of a single mobile light show device in the collaborative optimization algorithm, which enables each mobile light show device to have the ability to perform independent reasoning based on local information and participate in global collaboration, thereby realizing decentralized dynamic task allocation and behavior coordination.

[0035] In specific implementation, the determination of multi-dimensional state parameters for each particle based on the state dataset can be achieved in the following way: For each particle, the pose and energy state information of the corresponding mobile light show device are extracted from the state dataset. Then, a structured parameter description framework is designed to organize the pose and energy state information into a fixed-dimensional vector. This vector sequentially includes three-dimensional spatial coordinates and three-dimensional attitude angles representing the spatial state of the mobile light show device, linear acceleration and angular velocity representing the motion state, and remaining battery capacity and estimated remaining runtime representing energy consumption constraints. This vector is used as the multi-dimensional state parameters of the particle, thereby obtaining the multi-dimensional state parameters for each particle. The multi-dimensional state parameters are a digital representation of the particle's real-time physical state and resource constraints. They can provide accurate decision-making basis and environmental perception input for the cooperative optimization algorithm, ensuring that the generated cooperative strategy conforms to the actual motion capability and energy limit of the device, and avoiding unrealistic planning results.

[0036] In practical implementation, using the multi-dimensional state parameters of each particle as initial constraints, the decision population can be constructed based on the dynamic priorities of all light and shadow performance tasks in the following way: First, by performing a full permutation and combination of the key performance position points and adjustable lighting effect parameters associated with all atomic light and shadow performance tasks, a high-dimensional combination space containing all performance task sequences is constructed as a huge solution space for subsequent optimization search. Second, by performing reachability analysis on the three-dimensional spatial coordinates in the multi-dimensional state parameters of each particle and evaluating the load capacity of the remaining electrical capacity, the spatial range that the corresponding particle can effectively cover at the current moment and the energy consumption boundary for sustainably executing performance tasks are defined as the basic physical constraints for population initialization. Then, by introducing the dynamic priority values ​​of all light and shadow performance tasks, these priority values ​​are normalized and transformed into a probability distribution map that is non-uniformly distributed across the entire solution space. This process assigns higher weights to high-priority tasks and their associated performance schemes in the probability distribution map. Next, a hybrid initialization strategy combining random exploration and deterministic guidance is implemented across the vast solution space. This involves heuristically sampling the task sequence based on fundamental physical constraints and the probability distribution map to increase the probability of sampling high-quality solutions. Simultaneously, the specific three-dimensional spatial coordinates of each particle are combined, and a kinematic-based fast path generation method is used to construct an initial feasible movement trajectory from the starting point to the first sampled task point. Finally, by performing heuristic task sampling and path generation operations on each particle, a complete candidate solution vector encoding the initial task sequence, the corresponding trajectory path point sequence, and basic lighting parameters is assigned to it. The candidate solution vectors of all particles are then aggregated to form an initial solution set, thus completing the construction of the entire decision population. This decision population serves as the starting point for subsequent adaptive particle swarm optimization algorithms to iteratively search and improve performance.

[0037] It should be noted that in this application, the decision population is a global search and optimization space composed of all particles and their corresponding candidate cooperative strategies. Based on the decision population, the simulation of group parallel exploration can efficiently search the huge strategy combination space, thereby systematically finding the global cooperative scheme with the best overall performance, rather than local optimization.

[0038] In some embodiments, reference Figure 3 The figure is a flowchart illustrating the operation of determining a collaborative control strategy according to some embodiments of this application. In this application, the decision population is adaptively searched and iteratively solved using particle swarm optimization with switching modes based on the state dataset and the dynamic priority of each light show task. The collaborative control strategy for all movable light show devices to perform light shows can be implemented using the following steps: A multi-objective optimization function for collaborative control of lighting performances is constructed based on the state dataset and the dynamic priority of each lighting performance task. Particle swarm optimization iteratively is performed based on the decision population and the multi-objective optimization function; For each iteration, the evolutionary factor is determined based on the population distribution characteristics of the particle swarm; Based on the evolutionary factors, the search modes are dynamically switched to update the multi-dimensional state parameters of each particle, thereby obtaining the state parameter set of the particle swarm, and then obtaining the state parameter set of the particle swarm after each iteration. Once the iteration meets the termination condition, the iteration terminates and a collaborative control strategy for all movable light show devices to perform a light show is generated based on the state parameter set after the iteration ends.

[0039] In specific implementation, the multi-objective optimization function for coordinated control of light shows, based on the state dataset and the dynamic priorities of each light show task, can be constructed in the following way: First, the instantaneous minimum relative distance between all movable light show devices in the current coordinated control strategy and the average curvature of the trajectories of all movable light show devices can be calculated. The instantaneous minimum relative distance and the average curvature are then weighted and summed according to a preset ratio, for example, the instantaneous minimum relative distance is 0.6 and the average curvature is 0.4, to obtain the motion coordination degree that quantifies group obstacle avoidance safety and motion smoothness. Second, the estimated total mileage is obtained by accumulating the lengths of the planned path segments of all movable light show devices in the current coordinated control strategy. Simultaneously, the total system energy consumption estimate required to execute the strategy is calculated using a motion energy consumption model based on the total mileage and the remaining battery capacity of each movable light show device. Then, the light show tasks expected to be completed in the current coordinated control strategy are statistically analyzed, and their respective priorities are determined. The corresponding dynamic priority weights are weighted and accumulated to obtain the light and shadow narrative completion degree, which measures the efficiency of plan execution. Finally, the motion coordination degree, the estimated total system energy consumption, and the light and shadow narrative completion degree are normalized to their extreme values. Then, according to different real-time stages of the performance (e.g., opening, climax, and ending), different importance coefficients are assigned to the normalized motion coordination degree, the estimated total system energy consumption, and the light and shadow narrative completion degree through dynamic weight allocation technology. These are then linearly weighted and summed to obtain a scalarized comprehensive score, which serves as the specific output value of the multi-objective optimization function. The formula for calculating the comprehensive score is the multi-objective optimization function. The multi-objective optimization function is a mathematical evaluation tool used to comprehensively quantify and weigh multiple system performance indicators, such as motion coordination degree, estimated total system energy consumption, and the light and shadow narrative completion degree. Its function output value can serve as the direct target basis for driving and evaluating the iterative optimization of the particle swarm algorithm, guiding the search algorithm to select the overall optimal collaborative control strategy from numerous candidate solutions.

[0040] In specific implementation, particle swarm optimization iteration based on the decision population and the multi-objective optimization function can be implemented in the following way: Particle swarm optimization iteration is performed based on the decision population and the multi-objective optimization function. In each iteration, the multi-dimensional state vector of each particle is decoded into a specific device task allocation and path planning scheme, and substituted into the multi-objective optimization function for calculation, thereby updating the individual historical optimal solution of each particle and the global historical optimal solution of the entire population, providing a basis for subsequent velocity and position updates.

[0041] In specific implementation, the evolutionary factor can be determined based on the population distribution characteristics of the particle swarm in the following way: the average value of the Euclidean distance between particles can be calculated and normalized to obtain a value representing the degree of population aggregation as the evolutionary factor; wherein, the evolutionary factor is a quantitative indicator that dynamically represents the diversity state of the particle swarm aggregation degree. It can detect the risk of the algorithm getting stuck in local optima in real time and serve as the basis for triggering different search modes, thereby significantly improving the algorithm's global exploration and convergence ability in complex dynamic environments.

[0042] In specific implementation, the search modes are dynamically switched based on the evolutionary factor to update the multi-dimensional state parameters of each particle. The resulting state parameter set of the particle swarm can be achieved in the following way: First, the system predefines four search modes: convergence, development, exploration, and exit. Each mode corresponds to a different combination of acceleration coefficients and a learning object selection strategy. The corresponding search mode is dynamically determined and switched to by comparing the evolutionary factor with a preset threshold range: When the evolutionary factor value falls into a low range (e.g., 0-0.3), the convergence mode is used, focusing on learning from the individual's historical best and the global historical best; when the value falls into a low-to-medium range (e.g., 0.3-0.5), the development mode is used, focusing on learning from the individual's historical best and randomly selected excellent particles in the neighborhood; when the value falls into a high-to-medium range (e.g., 0.5-0.7), the exploration mode is used, focusing on learning from the individual's historical best and randomly selected excellent particles in the neighborhood. The algorithm learns from random particles within the optimal and globally optimal neighborhoods. When the value falls into the highest range (e.g., 0.7-1), it adopts an exit mode, focusing on learning from random neighboring particles and introducing larger random perturbations. Then, it synchronously configures and adjusts the individual learning acceleration coefficient, social learning acceleration coefficient, and target position to be learned (individual optimal position, neighborhood optimal position, or random perturbation direction) in the velocity update formula of each particle. Through the configured velocity update formula and position update formula, it iteratively calculates and refreshes the multidimensional state vector of each particle, thereby obtaining the latest multidimensional state vector of all particles after this round of iteration. The state parameter set is a set that records the multidimensional state vectors of all particles in the decision population during the iteration process. It can completely preserve the intermediate solution space of each iteration of the algorithm, providing a direct data foundation for the continuous optimization of the strategy and the decoding of the final strategy.

[0043] It should be noted that in this application, the collaborative control strategy is a complete set of instructions that specifies the specific actions and light effect outputs of all mobile lighting devices in the spatiotemporal dimension. It can transform offline intelligent computing results into an online, precisely executable collaborative performance blueprint, and is the direct technical carrier for realizing dynamic and adaptive group light show performances. In specific implementation, when the iteration meets the termination condition, the collaborative control strategy for terminating the iteration and decoding and generating all mobile light show devices for light show performance based on the state parameter set after the iteration ends can be implemented in the following way: when the number of iterations reaches a preset upper limit (e.g., 100 times) or the improvement of the optimal solution is less than 100 times, the collaborative control strategy for all mobile light show devices for light show performance can be implemented in the following way: When a preset threshold is reached (e.g., the difference between the output values ​​of the multi-objective optimization function in two consecutive iterations is less than 0.1), the termination condition is met and the optimization iteration process is stopped. The set of state parameters corresponding to the globally optimal particle generated in the final iteration is converted into specific instructions that can be executed by all movable light show devices through mapping decoding technology. That is, the abstract parameter sequence encoded in the multidimensional state vector of the particle is mapped one by one to the target position queue and light effect parameter sequence of each device. Thus, the set of all target position queues and light effect parameter sequences is used as the collaborative control strategy for all movable light show devices to perform light shows.

[0044] In step 104, the execution instruction set of each mobile light show device is determined according to the collaborative control strategy, and then each mobile light show device is driven to perform an adaptive and collaborative group light show within the performance area based on all the execution instruction sets.

[0045] In some embodiments, determining the set of execution instructions for each movable light show device according to the cooperative control strategy can be achieved by the following steps: For each movable light show device, the motion control command of the movable light show device is determined according to the target position queue and motion speed curve of the movable light show device in the cooperative control strategy. The light drive command for the movable light show device is generated based on the sequence of light effect parameters of the movable light show device in the collaborative control strategy. A unified high-precision time synchronization signal is injected into the motion control command and the light driving command, and packaged to generate an execution command set for the mobile light show device, thereby obtaining the execution command set for each mobile light show device.

[0046] In specific implementation, the motion control commands for the movable light show device, determined based on the target position queue and velocity curve of the movable light show device in the cooperative control strategy, can be implemented in the following way: First, spline interpolation is performed on the target position queue (i.e., a series of discrete three-dimensional spatial coordinate points) specified for the movable light show device in the cooperative control strategy to obtain a smooth spatial motion trajectory curve that satisfies continuous high-order derivative constraints. Then, based on the kinematic model of the movable light show device (such as a differential drive or omnidirectional movement model), inverse kinematics calculation is performed on the spatial motion trajectory curve to convert the desired pose and velocity of each point on the trajectory into the rotational speeds corresponding to the underlying independent drive units (such as motors). The calculated speed control and angle control quantities are then discretized and quantized according to a preset control cycle. The data is then encapsulated and verified according to a communication protocol format to obtain a discrete control data stream. This discrete control data stream is then converted into a pulse-width modulation signal waveform sequence that can be directly parsed by the corresponding hardware interface, serving as the motion control command for the mobile light show device. The motion control command is the underlying control signal sequence that directly drives the mobile light show device to achieve precise pose positioning and motion trajectory tracking during the light show performance. It can transform path planning into reliable mechanical motion, ensuring the safe, smooth, and coordinated movement of the mobile light show device group in complex environments.

[0047] In specific implementation, generating the lighting drive instructions for the movable light show device based on the lighting effect parameter sequence of the movable light show device in the collaborative control strategy can be achieved in the following way: First, extract the basic light effect type code and corresponding control parameters from the lighting effect parameter sequence specified for the movable light show device in the collaborative control strategy. Then, obtain the standard effect template and default parameters corresponding to the basic light effect type code by querying the preset lighting effect template library. Finally, use the embedded real-time rendering engine to parse and dynamically enhance the control parameters and default parameters to generate a more delicate light and color transition curve that conforms to human visual perception. Brightness gradient curves and dynamic pattern transformation sequences are used to obtain high-fidelity final light effect data. Finally, the final light effect data is formatted, encoded, and encapsulated according to industry-standard digital dimming protocols to generate a series of structured data packets with time stamp information. This results in light driving instructions that directly control the luminous state, color, and pattern of the light source, serving as the light driving instructions for the mobile light show device. The light driving instructions are digital control signals that dynamically control the color, intensity, and pattern of the light source to present a predetermined artistic effect. They can transform abstract light and shadow ideas into commands that the device can execute accurately, achieving complex, dynamic, and motion-synchronized light and shadow performances.

[0048] It should be noted that in this application, the execution instruction set is a final instruction package that can be independently and completely executed by a single mobile light show device. It can ensure that the actions and lighting effects of all devices are highly unified under strict time synchronization, which is the key to achieving global collaborative performance. In specific implementation, injecting a unified high-precision time synchronization signal into the motion control command and the light drive command, and packaging them to generate the execution instruction set of the mobile light show device can be achieved in the following way: the motion control command and the light drive command can be uniformly timed by a high-precision master clock source, and the two command streams carrying precise timestamps are aligned and sorted based on the time axis to ensure that the motion and light commands for the same moment are completely matched in time sequence, and the time-aligned command data is obtained; then, the command data is merged, encapsulated, and verification information such as cyclic redundancy check code is added to enhance transmission reliability, generating a unique and structured binary data packet set as the execution instruction set of the mobile light show device.

[0049] In some embodiments, driving each movable light show device to perform an adaptive and coordinated group light show within the performance area based on all execution instruction sets can be achieved through the following steps: All execution instruction sets are distributed to the device controllers of each mobile light show device through a distributed synchronous execution architecture. The controllers of each movable light show device execute motion servoing and light flux output in parallel according to the received execution instruction set, so as to realize adaptive and coordinated group light show.

[0050] In practical implementation, the distributed synchronous execution architecture can be used to distribute all execution instruction sets to the device controllers of each mobile light show device in the following way: First, the execution instruction set of each mobile light show device can be encoded and encapsulated through a low-latency wireless communication network between the central scheduling node and the device controllers to generate a structured execution instruction set with time stamp information. Then, a high-precision global clock synchronization protocol is introduced to synchronize the time between the central scheduling node and all device controllers, ensuring that the time base of the entire system is unified. Then, the central scheduling node distributes each structured execution instruction set point-to-point to the device controller of the corresponding mobile light show device through a time-triggered communication mechanism according to the unified performance timeline, and receives confirmation signals. At the same time, dynamic retransmission is used to compensate for and smooth out instruction packet loss or delay caused by network fluctuations, thereby reliably distributing accurate and synchronized execution instruction sets to the device controllers of each mobile light show device, completing the synchronous distribution of instructions for distributed collaborative tasks.

[0051] In practical implementation, the device controllers of each mobile light show device execute motion servoing and luminous flux output in parallel according to the received execution instruction set. The adaptive and coordinated group light show can be achieved in the following way: For each mobile light show device, after receiving the synchronously issued execution instruction set, the device controller immediately starts the local real-time control loop to decode and parse the instruction set, separating the target position, target posture, target speed, and target light parameters in the discrete time series. Then, the path tracking algorithm of the embedded kinematic model is called to perform local optimization and smoothing of the target trajectory according to the target position, target posture, and current real-time pose in the discrete time series, generating the pulse width modulation signal of the direct drive motor servo system. At the same time, the light rendering engine is called to generate the corresponding luminous flux output control signal according to the target light parameters in the discrete time series and the system timestamp. Thus, the decoded execution instruction set is transformed into precise physical motion and light and shadow output, realizing a stable, reliable, and adaptive coordinated group light show.

[0052] Furthermore, in another aspect of this application, in some embodiments, this application provides a portable cultural tourism nighttime light show control device, which includes a control unit, as referenced. Figure 4 The figure is a schematic diagram of the structure of a movable cultural tourism nighttime light show control device according to some embodiments of this application. The movable cultural tourism nighttime light show control device 400 includes: an acquisition module 401, a perception and decision-making module 402, and an execution module 403, which are described below: The acquisition module 401 in this application is mainly used to acquire the pose and energy state information of each movable light show device in the light device group, and then obtain the state dataset of the light device group. The perception and decision module 402 in this application is mainly used to perform point cloud three-dimensional modeling of the target cultural tourism night tour area, obtain a three-dimensional point cloud digital model of the performance area, and then determine the dynamic priority of each light and shadow performance task in the performance area based on the three-dimensional point cloud digital model and the preset performance plot timeline. It should be noted that the perception and decision module 402 in this application is also used to abstract each mobile light show device as a particle, and then construct a decision population based on all particles, and perform adaptive search mode switching particle swarm optimization iterative solution on the decision population through the state dataset and the dynamic priority of each light and shadow performance task, so as to obtain the collaborative control strategy for all mobile light show devices to perform light performances. The execution module 403 in this application is mainly used to determine the execution instruction set of each movable light show device according to the collaborative control strategy, and then drive each movable light show device to perform an adaptive and collaborative group light show in the performance area based on all the execution instruction sets.

[0053] The various modules in the aforementioned portable cultural tourism nighttime light show control device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0054] In another embodiment, this application provides a computer device, which may be a server, and its internal structure diagram may be as follows. Figure 5 As shown, the computer device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores data related to a mobile light show control method based on dynamic group collaboration. The network interface communicates with external terminals via a network connection. When executed by the processor, the computer program implements a mobile light show control method based on dynamic group collaboration.

[0055] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0056] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above embodiment of the mobile light show control method based on dynamic group collaboration.

[0057] In one embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps described in the embodiment of the mobile light show control method based on dynamic group collaboration.

[0058] In one embodiment, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the steps described in the embodiment of the mobile light show control method based on dynamic group collaboration.

[0059] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0060] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0061] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A control method for a mobile light show based on dynamic group collaboration, used to control mobile cultural tourism nighttime light show devices to perform dynamic light and shadow narrative performances in a group collaboration manner, characterized in that, The method includes the following steps: The pose and energy state information of each movable light show device in the light installation group are obtained, and then the state dataset of the light installation group is obtained. A point cloud 3D model is performed on the target cultural tourism night tour area to obtain a 3D point cloud digital model of the performance area. Then, based on the 3D point cloud digital model and the preset performance plot timeline, the dynamic priority of each light and shadow performance task in the performance area is determined. Each mobile light show device is abstracted as a particle, and a decision population is constructed based on all particles. The decision population is then subjected to adaptive search mode switching particle swarm optimization iterative solution through the state dataset and the dynamic priority of each light and shadow performance task to obtain the collaborative control strategy for all mobile light show devices to perform light performances. The execution instruction set for each mobile light show device is determined according to the collaborative control strategy, and then each mobile light show device is driven to perform an adaptive and collaborative group light show within the performance area based on all the execution instruction sets.

2. The method as described in claim 1, characterized in that, A 3D point cloud model of the target cultural tourism night tour area was created, resulting in a 3D point cloud digital model of the performance area, specifically including: Multi-view scanning of the target cultural tourism night tour area yielded the original three-dimensional point cloud data; A dense, colored point cloud covering the entire performance area is generated based on the original 3D point cloud data. Key environmental elements are identified in the dense colored point cloud, thereby constructing a three-dimensional point cloud digital model of the performance area.

3. The method as described in claim 1, characterized in that, Determining the dynamic priority of each light and shadow performance task in the performance area based on the aforementioned 3D point cloud digital model and the preset performance plot timeline specifically includes: The walkable region segmentation and visual focus extraction are performed on the three-dimensional point cloud digital model to obtain the key performance location point set of the performance area; Based on the preset performance plot timeline and the set of key performance locations, determine multiple light and shadow performance tasks associated with key performance locations; The urgency of each light and shadow performance task in terms of time is determined by identifying the key performance location points associated with all light and shadow performance tasks. Based on the urgency of all needs, tasks are sorted to obtain the dynamic priority of each light and shadow performance task in the performance area.

4. The method as described in claim 1, characterized in that, Each movable light show device is abstracted as a particle, and a decision population is constructed based on all particles, specifically including: Each movable light show installation is abstracted as a particle; Based on the state dataset, determine the multi-dimensional state parameters of each particle; Using the multi-dimensional state parameters of each particle as initial constraints, a decision population is constructed based on the dynamic priority of all light and shadow performance tasks.

5. The method as described in claim 1, characterized in that, By using the state dataset and the dynamic priorities of each light and shadow performance task, an adaptive search mode switching particle swarm optimization iterative solution is performed on the decision population to obtain the collaborative control strategy for all movable light show devices to perform light shows. Specifically, this strategy includes: A multi-objective optimization function for collaborative control of lighting performances is constructed based on the state dataset and the dynamic priority of each lighting performance task. Particle swarm optimization iteratively is performed based on the decision population and the multi-objective optimization function; For each iteration, the evolutionary factor is determined based on the population distribution characteristics of the particle swarm; Based on the evolutionary factors, the search modes are dynamically switched to update the multi-dimensional state parameters of each particle, thereby obtaining the state parameter set of the particle swarm, and then obtaining the state parameter set of the particle swarm after each iteration. Once the iteration meets the termination condition, the iteration terminates and a collaborative control strategy for all movable light show devices to perform a light show is generated based on the state parameter set after the iteration ends.

6. The method as described in claim 1, characterized in that, The execution instruction set for each movable light show device, determined according to the aforementioned collaborative control strategy, specifically includes: For each movable light show device, the motion control command of the movable light show device is determined according to the target position queue and motion speed curve of the movable light show device in the cooperative control strategy. The light drive command for the movable light show device is generated based on the sequence of light effect parameters of the movable light show device in the collaborative control strategy. A unified high-precision time synchronization signal is injected into the motion control command and the light driving command, and packaged to generate an execution command set for the mobile light show device, thereby obtaining the execution command set for each mobile light show device.

7. The method as described in claim 1, characterized in that, The group light show, which drives each movable light show device to perform adaptive and coordinated group light shows within the performance area based on all execution instruction sets, specifically includes: All execution instruction sets are distributed to the device controllers of each mobile light show device through a distributed synchronous execution architecture. The controllers of each movable light show device execute motion servoing and light flux output in parallel according to the received execution instruction set, so as to realize adaptive and coordinated group light show.

8. A portable control device for cultural tourism nighttime light shows, characterized in that, The device includes: The acquisition module is used to acquire the pose and energy state information of each movable light show device in the light installation group, thereby obtaining the state dataset of the light installation group. The perception and decision-making module is used to perform point cloud 3D modeling of the target cultural tourism night tour area to obtain a 3D point cloud digital model of the performance area, and then determine the dynamic priority of each light and shadow performance task in the performance area based on the 3D point cloud digital model and the preset performance plot timeline. The perception and decision-making module is used to abstract each mobile light show device as a particle, and then construct a decision population based on all particles. The decision population is then optimized iteratively by adaptively searching the state dataset and the dynamic priority of each light and shadow performance task to obtain the collaborative control strategy for all mobile light show devices to perform light performances. The execution module is used to determine the execution instruction set of each mobile light show device according to the collaborative control strategy, and then drive each mobile light show device to perform an adaptive and collaborative group light show within the performance area based on all the execution instruction sets.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the mobile light show control method based on dynamic group collaboration as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the mobile light show control method based on dynamic group collaboration as described in any one of claims 1 to 7.

Images