Robot target searching method used in perceptual restricted environment

Abstract

The invention discloses a robot target search method used in a perceptual restricted environment, and relates to the technical field of robot search, and the method comprises the steps: judging whether a signal sent by a current target is detected or not; if the signal sent by the current target is not detected, the robot is controlled to alternately walk in a Z shape, and the step of judging whether the signal sent by the current target is detected or not is executed after the robot walks one step every time until any one of the environment boundary reaching, task failure or the signal sent by the current target is detected is met; if the signal sent by the current target is detected, the robot is controlled to walk towards the current target until the robot walks to the position of the current target, and then searching of the current target is completed. Through Z-shaped alternate walking, the robot rapidly walks out of the boundary and bypasses the obstacle, invalid search can be reduced, and the search efficiency is improved.

Description

Technical Field

[0001] The present application relates to the field of robot search technology, and in particular to a robot target search method in a perception-constrained environment. Background Art

[0002] Multi-target search tasks in unknown environments are a key topic in robotics and intelligent agent research, with widespread applications in disaster relief, environmental monitoring, space exploration, and other fields. These scenarios place high demands on autonomous agents for efficient and intelligent target discovery and localization. For example, in disaster relief, agents must quickly locate survivors in unknown and complex debris fields; in environmental monitoring, agents are used to track and detect key points, such as pollution sources or biological habitats; and in space exploration, agents must locate key targets or resources in unexplored areas. Using robots to replace humans in search operations in dangerous environments not only reduces casualties but also improves mission efficiency, thus possessing significant practical significance. These tasks share common characteristics: the unknown environment, random target distribution, and the limited perception and mobility capabilities of the agents. These challenges contribute to the complexity and high demands of the search tasks. Summary of the Invention

[0003] In view of this, an embodiment of the present application provides a robot target search method for a perception-constrained environment, so as to improve the efficiency of the robot in target search in a perception-constrained environment.

[0004] An aspect of an embodiment of the present application provides a method for robot target search in a perception-constrained environment, the method comprising the following steps:

[0005] Determine whether the signal emitted by the current target is detected;

[0006] If the signal emitted by the current target is not detected, the robot is controlled to adopt a zigzag alternating walk, and the step of determining whether the signal emitted by the current target is detected is performed after each step of the robot until any one of reaching the environment boundary, mission failure, or detecting the signal emitted by the current target is satisfied;

[0007] If a signal sent by the current target is detected, the robot is controlled to walk toward the current target until the robot walks to the position of the current target, thereby completing the search for the current target.

[0008] In some embodiments, before determining whether a signal emitted by the current target is detected, the method further includes the following steps:

[0009] Acquire signals emitted by various targets in the environment;

[0010] Storing the signals sent by each of the targets in a cache queue;

[0011] Determine the target with the strongest signal in the cache queue as the current target;

[0012] The step of determining whether a signal emitted by the current target is detected comprises the following steps:

[0013] Determine whether a signal emitted by the current target is detected in the environment.

[0014] In some embodiments, after completing the search for the current target, the method further includes the following steps:

[0015] Determine whether there is a signal sent by other targets in the cache queue;

[0016] If there is a signal emitted by the other targets in the cache queue, the other targets that emit the strongest signal will be taken as the current target in turn, and then the control of the robot to walk towards the current target is returned until the robot walks to the position of the current target, thereby completing the step of searching for the current target, until the search of all targets in the cache queue is completed.

[0017] In some embodiments, after completing the search for the current target, the method further includes the following steps:

[0018] The signal sent by the current target and stored in the cache queue is deleted.

[0019] In some embodiments, controlling the robot to walk in a zigzag pattern and performing the step of determining whether a signal emitted by the current target is detected after each step of the robot until any one of reaching an environment boundary, mission failure, or detecting a signal emitted by the current target is satisfied includes the following steps:

[0020] If the robot is controlled to walk in a zigzag pattern and reaches the boundary of the environment, the confidence level of each area in the global environment is determined, and the robot is controlled to walk toward the area with the highest confidence level;

[0021] If the robot is controlled to walk in a zigzag pattern and the maximum number of iterations is reached, the task is judged to have failed;

[0022] If the robot is controlled to walk in a zigzag pattern and a signal from the current target is detected, the step of controlling the robot to walk toward the current target is executed until the robot walks to the position of the current target, thereby completing the step of searching for the current target.

[0023] In some embodiments, controlling the robot to walk toward the area with the highest confidence level comprises the following steps:

[0024] The robot is controlled to walk toward the center of mass of the area with the highest confidence.

[0025] In some embodiments, controlling the robot to walk toward the current target comprises the following steps:

[0026] A greedy strategy is adopted to control the robot to walk towards the current target.

[0027] Another aspect of the present application provides a robot target search device for sensing a restricted environment, the device comprising:

[0028] A signal detection unit, used to determine whether a signal emitted by the current target is detected;

[0029] an alternating walking unit, configured to control the robot to adopt a zigzag alternating walking if no signal from the current target is detected, and to execute the step of determining whether a signal from the current target is detected after each step of the robot until any one of reaching an environment boundary, mission failure, or detecting a signal from the current target is satisfied;

[0030] The target search unit is used to control the robot to walk towards the current target if a signal sent by the current target is detected, until the robot walks to the position of the current target, thereby completing the search for the current target.

[0031] Another aspect of the embodiments of the present application further provides an electronic device, including a processor and a memory;

[0032] The memory is used to store programs;

[0033] The processor executes the program to implement any of the above methods.

[0034] Another aspect of the embodiments of the present application further provides a computer-readable storage medium, wherein the storage medium stores a program, and the program is executed by a processor to implement any of the above methods.

[0035] This application has at least the following beneficial effects:

[0036] The present application can determine whether a signal from the current target is detected; if a signal from the current target is not detected, the robot is controlled to adopt a zigzag alternating walk, and after each step of the robot, the step of determining whether a signal from the current target is detected is executed until any one of the following conditions is met: reaching the environmental boundary, mission failure, or detecting a signal from the current target; if a signal from the current target is detected, the robot is controlled to walk toward the current target until the robot reaches the location of the current target, thereby completing the search for the current target. The present application uses zigzag alternating walks to allow the robot to quickly move out of the boundary and around obstacles, which can reduce invalid searches and improve search efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0038] Figure 1 A schematic diagram of an optional robot search direction provided in an embodiment of the present application;

[0039] Figure 2 A schematic diagram of a signal emitted by a target provided in an embodiment of the present application;

[0040] Figure 3 A flowchart of a method for searching a target in a constrained environment provided by an embodiment of the present application;

[0041] Figure 4 An example flow chart of a method for searching a target in a constrained environment provided by an embodiment of the present application;

[0042] Figure 5 This is an example effect diagram of the first solution in some optional implementations provided in the embodiments of the present application;

[0043] Figure 6 This is an example diagram of a robot walking after adding a random offset according to an embodiment of the present application;

[0044] Figure 7 This is an example diagram of the robot walking after adaptively adjusting its path according to an embodiment of the present application;

[0045] Figure 8 This is an example effect diagram of the second solution in some optional implementations provided in the embodiments of this application;

[0046] Figure 9 This is an example effect diagram of the third solution in some optional implementations provided in the embodiments of the present application;

[0047] Figure 10 This is a structural block diagram of a robot target search device for sensing a constrained environment provided in an embodiment of the present application. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solutions and advantages of this application more clearly understood, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0049] Before describing the embodiments of the present application in detail, some of the related technologies involved in the embodiments of the present application are first described as follows:

[0050] Explanation of terms:

[0051] Local Perception: In the field of robotic target search, local perception refers to the ability of a robot to acquire local environmental information through its sensors (such as cameras and lidar), and to understand and make decisions based on this information. It is one of the core technologies for autonomous navigation, target detection, and tracking, and is particularly important in dynamic or complex environments. Local perception emphasizes that the robot perceives and analyzes only the local environment within the current sensor coverage area, rather than modeling the global environment. Its core idea is to reduce computational complexity and improve real-time performance and adaptability by focusing on information extraction in a local area. Simply put, the robot can only perceive information within a certain range of its current location.

[0052] A* Algorithm: The A* algorithm is a heuristic search algorithm used to find the shortest path from a starting point to a destination in a weighted graph. It combines the advantages of breadth-first search (BFS) and greedy algorithms, efficiently guiding the search direction through a cost estimation function. It is widely used in path planning, game AI, robotic navigation, and other fields.

[0053] Breadth-first search (BFS) and depth-first search (DFS): Two different search algorithms.

[0054] Dijkstra: Same as above.

[0055] Artificial potential field method (APF): Same as above.

[0056] Path planning: Path planning is one of the main research areas of motion planning. Motion planning consists of path planning and trajectory planning. The sequence of points or curves connecting the starting and ending points is called a path, and the strategy that forms the path is called path planning.

[0057] Robot: The robot in this application has basic perception and positioning capabilities. Specifically, the sensor system equipped by the robot can accurately locate its own position, detect the intensity of the target signal, and perceive the geometric information of obstacles in the local range. These perception capabilities constitute the information basis for the intelligent agent to explore the environment and make decisions. However, due to the limitations of the sensor's perception range, the intelligent agent can only obtain environmental information within a limited area around it at any one time. This local observability is a constraint that needs to be considered in task design. The robot's action space and perception diagram are shown in the figure. The robot's action space is in four directions: up, down, left, and right, and the detection range is a 5x5 matrix. For example, Figure 1 A schematic diagram of an optional robot search direction.

[0058] Target: Targets are static and randomly distributed in the environment. This method assumes that the signal emitted by the target can be detected and recognized by the robot. Its intensity decreases with increasing distance. The maximum radiation range is D. When the robot walks to the target position, it means that the target has been found and collected by the robot, and thereafter the target will no longer emit signals into the environment. For example, Figure 2 Schematic diagram of the signal sent to the target.

[0059] Generally speaking, in search tasks, the environment of interest is a large, unknown environment with randomly distributed targets. Robots use sensors such as thermal imaging, vision, and sonar to perceive the targets' fitness values ​​and choose movement directions based on these values. In these application scenarios, intelligent agents often face a variety of complex environmental conditions, such as terrain obstacles, dynamic environmental changes, limited perception range, and restricted communication capabilities.

[0060] Traditional methods based on global path planning exhibit significant limitations when faced with unknown or dynamically changing environments. This problem is particularly prominent in high-dimensional random scenarios, mainly manifested in aspects such as excessive computational complexity and insufficient environmental adaptability. Although classic global information planning algorithms such as A* and Dijkstra can achieve optimal path solutions in deterministic static environments, their strict dependence on complete environmental information restricts their practical application efficiency in dynamic environments. In addition, as the complexity of the environment increases, the computational cost of the algorithm increases exponentially, making it difficult to meet real-time constraints. Although methods based on random sampling strategies have been widely studied in the field of path planning, their lack of systematic exploration mechanisms makes them prone to convergence to local optimal solutions, which significantly limits the applicability of such global methods in dynamic uncertain environments.

[0061] Compared to global planning methods, path planning strategies based on local perception demonstrate superior environmental adaptability and decision-making flexibility. Their core advantage lies in relying solely on the agent's local perception information for real-time path planning and decision updates. However, while the artificial potential field method (APF) performs well in complex obstacle environments, it is prone to falling into local optimal solutions and has difficulty ensuring global path optimality. While random search strategies based on Lévy flight and Gaussian random walks combined with simulated annealing algorithms demonstrate strong exploration capabilities, their limitations, limited to four discrete action spaces (up, down, left, and right), can easily lead to path loops, reducing search efficiency. In particular, in sparse environments with limited perception range, the agent can only obtain extremely limited local environmental information and lacks effective goal-oriented information, resulting in a significant decline in its global search performance.

[0062] The shortcomings of the existing technology and the technical problems to be solved by this application are:

[0063] In the research of object search tasks, existing methods can be roughly divided into three categories: traditional rule-based methods, heuristic-based random search methods, and learning-based intelligent methods. Each method has its own unique advantages and disadvantages, and performs differently in different environments and tasks.

[0064] Traditional rule-based methods, such as breadth-first search (BFS) and depth-first search (DFS), can systematically cover the entire search space to ensure that the target is found. However, these methods are often computationally expensive and lack adaptability when dealing with dynamic and high-dimensional environments, and cannot meet the real-time requirements in complex scenarios. In addition, potential field-based search strategies, such as the artificial potential field method (APF), guide the movement of the agent by defining the attraction and repulsion between the target and obstacles. They perform well in dealing with environments with dense obstacles, but APF is prone to falling into local optimality, resulting in the agent being unable to effectively find the global optimal path. To this end, many improved methods have been proposed, such as introducing random perturbations or combining other global search algorithms to improve the ability to escape from local optimal solutions.

[0065] Heuristic random search methods, such as Lévy flights and random walks, introduce randomness to improve search coverage and are suitable for environments with sparse information. These methods are advantageous in their simplicity and versatility, enabling effective exploration over a wide range, particularly when target signals are sparse. However, these methods lack precise goal-directedness and tend to lose efficiency as they approach the target. Therefore, they are often combined with other more directive strategies, such as potential field-based methods, to improve search precision and efficiency.

[0066] Intelligent optimization algorithms, such as particle swarm optimization (PSO) and simulated annealing (SA), can find optimal solutions in complex environments by simulating group behavior in nature or utilizing probabilistic search mechanisms. PSO accelerates convergence through information sharing between individuals and is particularly suitable for collaborative search tasks involving multiple agents. SA, on the other hand, escapes from local optimal solutions by gradually lowering the "temperature," making it suitable for global optimization in larger search spaces. These methods may lack flexibility in dynamic environments, but their adaptability can be enhanced when combined with other heuristic methods. For example, Related Technology 1 proposes a probabilistic finite state machine-based multi-objective search strategy (PFSMS) for swarm robots, and Related Technology 2 proposes a hierarchical path prediction method (HDMPC) with a trigger mechanism, both of which have demonstrated excellent performance in various scenarios.

[0067] Learning-based intelligent methods, such as deep reinforcement learning (DRL), have demonstrated great potential in object search tasks in recent years. Through continuous interaction with the environment, DRL can learn optimal search strategies in complex environments, making it particularly suitable for dynamic and unstructured environments. Compared to traditional methods, learning-based approaches are better able to cope with environmental uncertainty and dynamic changes. However, training DRL models typically requires extensive computational resources and samples, and faces challenges with generalization and robustness in the real world, which limits their practical application.

[0068] In summary, existing target search methods each have their own strengths and weaknesses. Rule-based methods, while ensuring systematicity, lack adaptability; heuristic methods are simple and efficient, but lack high precision; and intelligent learning-based methods can perform well in complex environments but are resource-intensive.

[0069] Reference Figure 3 , the embodiment of the present application provides a robot target search method for perceiving a restricted environment, specifically comprising the following steps S300~S320:

[0070] S300: Determine whether a signal from the current target is detected;

[0071] S310: If the signal emitted by the current target is not detected, controlling the robot to adopt a zigzag alternating walk, and executing the step of determining whether the signal emitted by the current target is detected after each step of the robot, until any one of reaching the environment boundary, mission failure, or detecting the signal emitted by the current target is satisfied;

[0072] S320: If a signal emitted by the current target is detected, the robot is controlled to walk toward the current target until the robot walks to the position of the current target, thereby completing the search for the current target.

[0073] Optionally, before determining whether a signal emitted by the current target is detected, the method further comprises the following steps:

[0074] Acquire signals emitted by various targets in the environment;

[0075] Storing the signals sent by each of the targets in a cache queue;

[0076] Determine the target with the strongest signal in the cache queue as the current target;

[0077] The step of determining whether a signal emitted by the current target is detected comprises the following steps:

[0078] Determine whether a signal emitted by the current target is detected in the environment.

[0079] Optionally, after completing the search for the current target, the method further comprises the following steps:

[0080] Determine whether there is a signal sent by other targets in the cache queue;

[0081] If there is a signal emitted by the other targets in the cache queue, the other targets that emit the strongest signal will be taken as the current target in turn, and then the control of the robot to walk towards the current target is returned until the robot walks to the position of the current target, thereby completing the step of searching for the current target, until the search of all targets in the cache queue is completed.

[0082] Optionally, after completing the search for the current target, the method further comprises the following steps:

[0083] The signal sent by the current target and stored in the cache queue is deleted.

[0084] Optionally, the step of controlling the robot to alternately walk in a zigzag pattern and performing the step of determining whether a signal emitted by the current target is detected after each step of the robot until any one of reaching an environment boundary, mission failure, or detecting a signal emitted by the current target is satisfied comprises the following steps:

[0085] If the robot is controlled to walk in a zigzag pattern and reaches the boundary of the environment, the confidence level of each area in the global environment is determined, and the robot is controlled to walk toward the area with the highest confidence level;

[0086] If the robot is controlled to walk in a zigzag pattern and the maximum number of iterations is reached, the task is judged to have failed;

[0087] If the robot is controlled to walk in a zigzag pattern and a signal from the current target is detected, the step of controlling the robot to walk toward the current target is executed until the robot walks to the position of the current target, thereby completing the step of searching for the current target.

[0088] Optionally, controlling the robot to walk toward the area with the highest confidence level comprises the following steps:

[0089] The robot is controlled to walk toward the center of mass of the area with the highest confidence.

[0090] Optionally, controlling the robot to walk toward the current target comprises the following steps:

[0091] A greedy strategy is adopted to control the robot to walk towards the current target.

[0092] Next, the solution of the embodiment of the present application will be introduced and explained in detail with reference to specific application examples.

[0093] Specifically, this embodiment includes the following technical solutions:

[0094] Solution 1: To overcome the limitations of traditional random walk strategies, where the discrete action space of up, down, left, and right easily leads to path loops, this embodiment proposes an exploration pattern based on alternating zigzag motion. This method constructs a systematic spatial coverage strategy through alternating horizontal and vertical motion, effectively avoiding repeated path exploration while achieving efficient traversal of the search space. This structured motion pattern has stronger directionality and spatial coverage than random walks, significantly improving the agent's exploration efficiency.

[0095] To ensure the robustness of the exploration strategy, this embodiment designs specific optimization mechanisms for boundary processing and obstacle avoidance:

[0096] 1.1. In order to avoid the path cycle problem when bouncing off the boundary, the algorithm introduces a random perturbation so that the agent can break the possible cycle path.

[0097] 1.2. When encountering an obstacle that makes it impossible to maintain the alternating motion pattern, the agent will adopt an adaptive adjustment strategy to restore the alternating motion pattern as quickly as possible while ensuring safety.

[0098] The second solution, to address the problem of target loss caused by the detection of other target signals during the target development process, this embodiment introduces a target information caching mechanism. This mechanism allows the agent to prioritize the currently tracked target when detecting multiple targets simultaneously, while temporarily storing characteristic information such as the location and intensity of other targets in a cache queue. This hierarchical processing strategy ensures that the agent can quickly switch to the next target to be processed based on the cached information after completing the current target development, effectively avoiding target information loss due to attention shifting, thereby improving the agent's search continuity and target development efficiency in multi-target environments.

[0099] The third solution, to achieve efficient global search capabilities, proposes an exploration strategy based on regional confidence assessment. This method first dynamically divides the search space into multiple unexplored blocks, and then quantitatively analyzes the potential value of each block by establishing a confidence evaluation mechanism. Based on this evaluation system, the agent selects the unexplored area with the highest confidence and uses its centroid as a guide point for path planning. This strategy, which combines regional segmentation, confidence assessment, and centroid guidance, gives the agent a global planning capability, achieving systematic exploration of the global search space while maintaining local decision-making flexibility, effectively improving search efficiency.

[0100] Next, this embodiment will be further described with reference to the accompanying drawings.

[0101] Reference Figure 4 , this embodiment provides an example flowchart of a method for robot target search in a perceptually constrained environment.

[0102] When the robot enters the target area, it initializes its movement direction based on its current location. If it is in the lower left corner, the initial direction is upper right. Similarly, if the robot is initially in the upper left corner of the map, the initial movement direction is lower right. During each iteration, the robot alternates between selecting a movement direction from a composite set of directions (right, down, right, down, etc.). Furthermore, after each movement, the robot updates its environmental awareness: a real-time perception (with a perception radius of R) and an environmental mapping matrix (which identifies known areas and locations for regional redirection). If a target signal is detected, a target signal caching mechanism intervenes, selecting the maximum signal and storing other detected target signals. The robot then gradually approaches the target center. Once the robot reaches the target location, the target is detected and marked as collected. This target no longer emits signals.

[0103] When the robot reaches the boundary, it will detect the known area and divide the entire area into multiple unknown areas. At this time, the redirection mechanism is used to guide the robot to the center of the most valuable area.

[0104] This process repeats until the maximum number of iterations is reached or all targets are collected.

[0105] Reference Figure 5 ,by Figure 5 The effect of the first solution of this embodiment is described below:

[0106] Map size: 10x10, robot starting position: (1,2), obstacle position: (8,6) (8,7).

[0107] The robot adopts the zigzag alternating motion proposed in this embodiment to expand the search area as much as possible and will not repeatedly explore a certain area like a random algorithm. Figure 5 The medium blue represents the walking path.

[0108] 1.1. However, this method may still result in a loop (a larger loop), leading to repetitive movement. Therefore, in this embodiment, 1.1 adds a random offset to 1 to impart a degree of randomness to the robot's movement. Instead of immediately bouncing upon reaching the boundary, the robot walks a random distance along the boundary before bouncing. The amount of the bounce is typically between two and four times the radius of the robot's detection range. Figure 6 The green color in the represents the offset path.

[0109] 1.2. In addition, when encountering an obstacle, the method proposed in this embodiment may not be able to move alternately normally. In this case, this embodiment allows the robot to continue moving in the current obstacle-free direction until it can move alternately, and then immediately change direction. Figure 7 In the figure, brown represents the adaptive adjustment path, and green represents the offset path.

[0110] Combining the previous basic methods and the improvements in 1.1 and 1.2, the robot can better search a larger area within the area, thereby improving its ability to search for targets.

[0111] by Figure 8 The effect of the second solution proposed in this embodiment is described below:

[0112] Map size: 20x20, robot starting position: (17,0), there are several obstacles in the map.

[0113] The robot enters the search area from coordinates (17,0) and explores the area in a zigzag pattern. When it reaches position (11,6), while tracking target A at (11,7), it also detects the signal from target B at (6,4). After the robot reaches (11,7) and completes its collection of target A, it can no longer detect target B's signal at this location. Thanks to the target information caching mechanism, the robot is able to quickly track target B again, completing the task quickly.

[0114] Simply put, when two or more target signals are detected at a certain location, the target signal strength and target number are stored. The robot then searches for the closest target (the one with the strongest signal) first. After collecting the target with the strongest signal, if the robot can sense signals from other targets, it will proceed directly to the signal location. If it cannot sense the target signal, it will check whether the target cache contains any information. If so, it will retrieve it and proceed to the location with the highest signal in the cache. After collecting the target, it will be deleted from the cache. If the target cache does not contain any information, the robot will continue exploring according to the current strategy.

[0115] by Figure 9 The effect of the third solution of this embodiment is described below:

[0116] Map size: 20x20. The robot starts at (20, 10) and walks to (9, 20). At this point, the robot divides the entire search area into two location regions for the known areas of the environment. Using the region confidence evaluation mechanism established in this embodiment, the value of region A is calculated to be 91, and the value of region B is calculated to be 21. At this point, this embodiment determines that the larger region is more likely to contain the target. Subsequently, the center of region A is calculated, and the direction of the center of region A relative to the robot's current position is used to guide the robot toward region A.

[0117] The beneficial effects of this embodiment include:

[0118] This embodiment addresses key issues such as path duplication, target loss, and global exploration efficiency in multi-target search tasks, and proposes a multi-target search algorithm (Scare) based on a zigzag exploration mode, information caching mechanism, and regional confidence guidance. The algorithm achieves innovation in the following three aspects: (1) It designs an exploration mode based on zigzag alternating movement, replacing the traditional four-way random walk strategy with structured movement in the horizontal and vertical directions, effectively avoiding the path loop problem and improving spatial coverage efficiency; (2) It constructs a target information caching mechanism, realizes hierarchical processing and orderly development of multi-target information, and solves the problem of target loss caused by the agent's attention shift during the target development process; (3) It proposes a global guidance strategy based on regional confidence evaluation, which enhances the algorithm's global search capability by combining dynamic region division, confidence quantification, and centroid guidance.

[0119] Reference Figure 10 , an embodiment of the present application provides a robot target search device for sensing in a constrained environment, comprising:

[0120] A signal detection unit, used to determine whether a signal emitted by the current target is detected;

[0121] an alternating walking unit, configured to control the robot to adopt a zigzag alternating walking if no signal from the current target is detected, and to execute the step of determining whether a signal from the current target is detected after each step of the robot until any one of reaching an environment boundary, mission failure, or detecting a signal from the current target is satisfied;

[0122] The target search unit is used to control the robot to walk towards the current target if a signal sent by the current target is detected, until the robot walks to the position of the current target, thereby completing the search for the current target.

[0123] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0124] In some optional embodiments, the functions / operations mentioned in the block diagram may not occur in the order mentioned in the operation diagram. For example, depending on the functions / operations involved, the two boxes shown in succession may actually be executed substantially simultaneously or the boxes may sometimes be executed in reverse order. In addition, the embodiments presented and described in the flow chart of the present application are provided in an exemplary manner for the purpose of providing a more comprehensive understanding of the technology. The disclosed method is not limited to the operations and logic flows presented in the present embodiment. Optional embodiments are contemplated in which the order of the various operations is changed and the sub-operations described as a part of a larger operation are performed independently.

[0125] In addition, although the present application is described in the context of functional modules, it should be understood that, unless otherwise stated, one or more of the functions and / or features described may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in separate physical devices or software modules. It is also understood that a detailed discussion of the actual implementation of each module is not necessary for understanding the present application. More specifically, given the properties, functions, and internal relationships of the various functional modules in the device disclosed in this embodiment, the actual implementation of the module will be understood within the routine skills of an engineer. Therefore, a person skilled in the art can implement the present application as set forth in the claims using ordinary techniques without undue experimentation. It is also understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present application, which is determined by the full scope of the appended claims and their equivalents.

[0126] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program code.

[0127] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0128] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, and then editing, interpreting, or processing in another suitable manner as necessary, and then storing it in a computer memory.

[0129] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having logic gate circuits for implementing logic functions on data signals, an application-specific integrated circuit having suitable combinational logic gate circuits, a programmable gate array (PGA), a field-programmable gate array (FPGA), etc.

[0130] Throughout this specification, reference to terms such as "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" means that a specific feature, structure, material, or characteristic described in conjunction with that embodiment or example is included in at least one embodiment or example of the present application. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0131] Although the embodiments of the present application have been shown and described, those skilled in the art will appreciate that various changes, modifications, substitutions, and variations may be made to the embodiments without departing from the principles and intent of the present application, and that the scope of the present application is defined by the claims and their equivalents.

[0132] The above is a specific description of the preferred implementation of the present application, but the present application is not limited to the embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present application, and these equivalent modifications or substitutions are all included in the scope defined by the claims of the present application.

Claims

1. A robot target search method for perceiving a constrained environment, characterized in that: The method comprises the following steps: Determine whether the signal emitted by the current target is detected; If the signal emitted by the current target is not detected, the robot is controlled to adopt a zigzag alternating walk, and the step of determining whether the signal emitted by the current target is detected is performed after each step of the robot until any one of reaching the environment boundary, mission failure, or detecting the signal emitted by the current target is satisfied; If a signal sent by the current target is detected, the robot is controlled to walk toward the current target until the robot walks to the position of the current target, thereby completing the search for the current target.

2. The method for robot target search in a constrained environment according to claim 1, characterized in that: Before determining whether a signal emitted by the current target is detected, the method further includes the following steps: Acquire signals emitted by various targets in the environment; Storing the signals sent by each of the targets in a cache queue; Determine the target with the strongest signal in the cache queue as the current target; The step of determining whether a signal emitted by the current target is detected comprises the following steps: Determine whether a signal emitted by the current target is detected in the environment.

3. The robot target search method for sensing a restricted environment according to claim 2, characterized in that: After completing the search for the current target, the method further includes the following steps: Determine whether there is a signal sent by other targets in the cache queue; If there is a signal emitted by the other targets in the cache queue, the other targets that emit the strongest signal will be taken as the current target in turn, and then the control of the robot to walk towards the current target is returned until the robot walks to the position of the current target, thereby completing the step of searching for the current target, until the search of all targets in the cache queue is completed.

4. The method for robot target search in a constrained environment according to claim 2, wherein: After completing the search for the current target, the method further includes the following steps: The signal sent by the current target and stored in the cache queue is deleted.

5. The method for robot target search in a constrained environment according to claim 1, characterized in that: The step of controlling the robot to alternately walk in a zigzag pattern and performing the step of determining whether a signal emitted by the current target is detected after each step of the robot until any one of reaching an environment boundary, mission failure, or detecting a signal emitted by the current target is satisfied comprises the following steps: If the robot is controlled to walk in a zigzag pattern and reaches the boundary of the environment, the confidence level of each area in the global environment is determined, and the robot is controlled to walk toward the area with the highest confidence level; If the robot is controlled to walk in a zigzag pattern and the maximum number of iterations is reached, the task is judged to have failed; If the robot is controlled to walk in a zigzag pattern and a signal from the current target is detected, the step of controlling the robot to walk toward the current target is executed until the robot walks to the position of the current target, thereby completing the step of searching for the current target.

6. The method for robot target search in a constrained environment according to claim 5, characterized in that: The controlling the robot to walk toward the area with the highest confidence level comprises the following steps: The robot is controlled to walk toward the center of mass of the area with the highest confidence.

7. The robot target search method for use in a perception-constrained environment according to any one of claims 1 to 6, characterized in that: The controlling the robot to walk toward the current target comprises the following steps: A greedy strategy is adopted to control the robot to walk towards the current target.

8. A robot target search device for sensing in a constrained environment, characterized in that: The device comprises: A signal detection unit, used to determine whether a signal emitted by the current target is detected; an alternating walking unit, configured to control the robot to adopt a zigzag alternating walking if no signal from the current target is detected, and to execute the step of determining whether a signal from the current target is detected after each step of the robot until any one of reaching an environment boundary, mission failure, or detecting a signal from the current target is satisfied; The target search unit is used to control the robot to walk towards the current target if a signal sent by the current target is detected, until the robot walks to the position of the current target, thereby completing the search for the current target.

9. An electronic device, characterized in that: The electronic device includes a processor and a memory; The memory is used to store programs; The processor executes the program to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The storage medium stores a program, and the program is executed by a processor to implement the method according to any one of claims 1 to 7.

Images