Path planning method for battery optimization and system supporting same

Abstract

A path planning system for battery optimization according to one embodiment of the present invention comprises: a map generation unit that generates a map including a charging station for a robot; a position setting unit that sets an initial position and a target position of the robot; a graph generation unit that generates a graph on the basis of a minimum coverage node; and a path planning unit that plans a path of the robot on the basis of a hierarchical heuristic algorithm.

Description

Path planning method for battery optimization and system supporting the same

[0001] The present invention relates to a path planning method for battery optimization and a system supporting the same.

[0002] Generally, driving methods for robots include line tracking, random driving, map-based driving, formation control, and coverage path planning. Among these, Coverage Path Planning (CPP) refers to the process of planning an optimal path so that a robot or autonomous vehicle (hereinafter referred to as a robot) can effectively move within a driving area. Various automated systems, such as cleaning robots, agricultural robots, and surveillance robots, can move autonomously using CPP.

[0003] Conventional CPPs are designed to minimize travel path length and optimize movement efficiency; however, this approach fails to consider critical constraints such as robot battery consumption, leading to the problem of the battery running out while the robot is moving along the CPP path. This issue can significantly reduce the efficiency of the overall system, particularly when multiple robots are operating in large-scale or complex environments.

[0004] The present invention aims to solve the problem described above by providing a path planning method for battery optimization and a system supporting the same, which derives a path capable of effectively covering the entire area while minimizing the robot's battery consumption by utilizing a hierarchical heuristic approach that uses ant colony optimization and genetic algorithms to monitor the battery status in real time and plan the optimal charging time and path.

[0005] The objectives of the present invention are not limited to those mentioned above, and other unmentioned objectives will be clearly understood by those skilled in the art from the description below.

[0006] A path planning system for battery optimization according to an embodiment of the present invention for achieving the above technical problem comprises a map generation unit that generates a map including a robot's charging station, a position setting unit that sets the robot's initial position and target position, a graph generation unit that generates a graph based on minimum coverage nodes, and a path planning unit that plans the robot's path based on a hierarchical heuristic algorithm, wherein the graph represents the robot's movement area.

[0007] In addition, the path planning unit can determine the robot's initial path sequence based on an ant colony optimization algorithm.

[0008] In addition, the path planning unit can determine the initial path sequence of the robot based on an ant colony optimization algorithm that adjusts pheromone intensity based on the robot's battery status.

[0009] In addition, the path planning unit can plan the robot's path by introducing redundant nodes into the determined initial path sequence and applying a genetic algorithm for final path optimization.

[0010] In addition, the path planning unit can plan the robot's path by applying a genetic algorithm to an initial path sequence in which at least one of crossover and mutation of the genetic algorithm is performed based on the robot's battery status.

[0011] In addition, a path planning system for battery optimization according to one embodiment of the present invention may further include a battery charging execution unit that causes the robot to arrive at a charging station and charge the battery based on the monitored battery status of the robot as it moves along the planned path of the robot.

[0012] A path planning method for battery optimization according to an embodiment of the present invention for achieving the above technical problem comprises the steps of: generating a map including a charging station of a robot; setting an initial position and a target position of a robot; generating a graph based on a minimum coverage node; and planning a path of a robot based on a hierarchical heuristic algorithm, wherein the graph represents the movement area of ​​the robot.

[0013] In addition, the step of planning the robot's path based on a hierarchical heuristic algorithm may include the step of determining the robot's initial path sequence based on an ant colony optimization algorithm.

[0014] Additionally, the step of determining the initial path sequence of the robot based on an ant colony optimization algorithm may include the step of determining the initial path sequence of the robot based on an ant colony optimization algorithm that adjusts the pheromone intensity based on the battery status of the robot.

[0015] Additionally, the step of planning the robot's path based on a hierarchical heuristic algorithm may include a step of determining the robot's initial path sequence based on an ant colony optimization algorithm and a step of planning the robot's path by introducing duplicate nodes to the determined initial path sequence and applying a genetic algorithm for final path optimization.

[0016] Additionally, the step of planning the robot's path by applying a genetic algorithm for introducing duplicate nodes and optimizing the final path to a determined initial path sequence may include the step of planning the robot's path by applying a genetic algorithm to the initial path sequence in which at least one of crossover and mutation of the genetic algorithm is performed based on the robot's battery state.

[0017] Additionally, the step of planning the robot's path by introducing redundant nodes into the determined initial path sequence and applying a genetic algorithm for final path optimization may include the step of monitoring the robot's battery status as the robot moves along the planned robot path and the step of arriving at a charging station and charging the battery based on the results of the monitoring.

[0018] According to the present invention, to achieve this, a hierarchical heuristic approach using ant colony optimization and genetic algorithms is utilized to monitor battery status in real time and plan optimal charging times and paths, thereby enabling effective coverage of the entire area while minimizing battery consumption of the robot.

[0019] In addition to this, various effects that can be identified directly or indirectly through this document may be provided.

[0020] FIG. 1 is a configuration diagram of a path planning system for battery optimization according to one embodiment of the invention.

[0021] Figure 2 is an example of map generation according to the prior art.

[0022] FIG. 3 is an exemplary diagram of a path planning system for battery optimization according to another embodiment of the invention.

[0023] FIG. 4 is a flowchart of a path planning method for battery optimization according to one embodiment of the invention.

[0024] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the attached drawings. The advantages and features of the present invention and the methods for achieving them will become clear by referring to the embodiments described below in detail together with the attached drawings. However, the present invention is not limited to the embodiments disclosed below but may be implemented in various different forms. These embodiments are provided merely to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the invention, and the present invention is defined only by the scope of the claims. In relation to the description of the drawings, the same or corresponding components may be assigned the same reference number.

[0025]

[0026] Although terms such as "first," "second," etc. are used to describe various elements, components, and / or sections, it goes without saying that these elements, components, and / or sections are not limited by these terms. These terms are used merely to distinguish one element, component, or section from another. Accordingly, it goes without saying that the first element, first component, or first section mentioned below may be a second element, second component, or second section within the technical scope of the present invention.

[0027] The terms used herein are for describing embodiments and are not intended to limit the invention. In this specification, the singular form includes the plural form unless specifically stated otherwise in the text. As used herein, "comprises" and / or "made of" do not exclude the presence or addition of one or more other components, steps, actions, and / or elements to the mentioned components, steps, actions, and / or elements.

[0028] Unless otherwise defined, all terms used in this specification (including technical and scientific terms) may be used in a meaning commonly understood by those skilled in the art to which the present invention pertains. Additionally, terms defined in commonly used dictionaries are not to be interpreted ideally or excessively unless explicitly and specifically defined otherwise.

[0029] Hereinafter, the configuration of the present invention will be described in detail with reference to the attached drawings.

[0030]

[0031] FIG. 1 is a configuration diagram of a path planning system for battery optimization according to one embodiment of the present invention.

[0032] Referring to FIG. 1, a path planning system (100) for battery optimization according to one embodiment of the present invention (hereinafter referred to as the battery optimization path planning system (100)) includes a map generating unit (110) that generates a map including a charging station of a robot, a position setting unit (120) that sets an initial position and a target position of a robot, a graph generating unit (130) that constructs a graph based on a minimum coverage node, and a path planning unit (140) that plans the path of the robot based on a hierarchical heuristic algorithm.

[0033] Additionally, a battery optimization path planning system (100) according to one embodiment may further include a battery charging execution unit that allows the robot to arrive at a charging station and charge the battery based on the monitored battery status of the robot when the robot moves along the planned path of the robot.

[0034] As mentioned above, when designing a CPP, which is a process of planning a path so that a robot can efficiently move through the coverage, which is all accessible space within a given area, it is necessary to design it by considering the following points in order to maximize efficiency by perfectly covering the designated space while simultaneously minimizing the length or time of the path.

[0035] (1) Power and battery management: Path planning that takes into account battery levels so that the robot does not run out of power while performing tasks or moving.

[0036] (2) Obstacle avoidance: Path planning to effectively avoid obstacles.

[0037] (3) Optimization: Planning a path by applying optimization techniques to minimize the length or time of the path.

[0038] (4) Dynamic environment adaptation: Path planning to quickly adapt to changing environments and replan the path.

[0039] A battery optimization path planning system (100) according to one embodiment plans a robot's path based on a hierarchical heuristic algorithm. A heuristic algorithm is an algorithm that uses empirical techniques or practical methods, and focuses on quickly finding an approximate or good solution through practical methods rather than finding a perfect or absolute solution to a problem.

[0040] Hierarchical heuristic algorithms apply various heuristic algorithms across multiple layers or levels to solve a problem; each layer utilizes a different algorithm to address a specific part of the problem while progressively advancing toward the overall solution. Using hierarchical heuristic algorithms allows for the flexible selection and application of the algorithm best suited to a particular aspect of the problem. Furthermore, by specializing algorithms at each layer, the efficiency of overall problem solving can be enhanced, and different algorithms can be flexibly applied to various problem situations.

[0041] For example, a path planning unit (140) according to one embodiment may use a hierarchical heuristic algorithm to select and apply the algorithm most suitable for a specific problem or situation, and as described below, in the initial stage, it may perform fast path search using Ant Colony Optimization (ACO), and in the subsequent stage, it may further optimize this path through Genetic Algorithms (GA).

[0042] Ant colony optimization is a heuristic algorithm that mimics the behavior of natural ants in finding optimal paths by utilizing the pheromone trails they leave behind while searching for food. The basic operating principle of ACO is as follows.

[0043] (1) Pheromone initialization: Initially, a small amount of pheromone is applied to all paths. This allows the ants to explore the paths randomly.

[0044] (2) Ant Pathfinding: Each ant starts from a starting point and searches for a path to a target point. The ant selects the next node at each step by considering the amount of pheromones and heuristic information (e.g., the reciprocal of the distance). Generally, it prefers paths with a large amount of pheromones and a short distance.

[0045] (3) Pheromone Update: Once all ants have completed a path, the amount of pheromone added to each path is determined based on the quality of the path found by the ants. More pheromones are added to shorter or lower-cost paths, and the amount of pheromone evaporation can also be determined so that outdated path information gradually disappears during the optimization process over time.

[0046] (4) Repetition: As the process of (1) to (3) above is repeated several times, the ant gradually evolves toward the optimal path. At this time, as time passes, pheromones accumulate intensively along the optimal path, and the frequency of exploration of other paths decreases relatively.

[0047] Genetic algorithms are heuristic algorithms applicable to solving optimization problems such as search and path optimization; they mimic genetic operations such as selection, crossover, and mutation that occur during the process of biological evolution. GAs are suitable for parallel processing that evaluates multiple solutions simultaneously and can search for global optima with high probability without getting stuck on local optima. The basic operating principles of GA are as follows.

[0048] (1) Encoding: Encodes the solution to the problem into a single string of characters, called a chromosome, that represents the gene. Generally, binary numbers, real numbers, or ordered lists can be used.

[0049] (2) Initial population: Starts with an initial set of randomly generated chromosomes. This initial set is called the population, and each chromosome represents a potential solution to the problem.

[0050] (3) Fitness function: A function that evaluates the quality of each chromosome, which numerically evaluates the performance of each solution for a given problem to determine how suitable the solution is.

[0051] (4) Selection: One or two chromosomes (parents) are selected from the current population to be used as parents for the next generation. At this time, the more suitable the chromosome, the higher the probability of being selected.

[0052] (5) Crossover: Genetic information from selected parent chromosomes is combined to create a new chromosome (offspring). At this time, some of the parents' genes are exchanged.

[0053] (6) Mutation: Maintains diversity by introducing random small changes to offspring chromosomes. Although the probability of mutation is generally very low, it can introduce new genetic variations and expand the search space.

[0054] (7) Repeat: Form a new population with the generated children, and repeat this process.

[0055] A battery optimization path planning system (100) according to one embodiment uses GA to generate an optimal path so that a robot can visit a charging station while efficiently covering each node. At this time, each solution within the population represents a path including a visit to a charging station, and these paths evolve in a direction that minimizes energy consumption and shortens the time to complete the goal. In addition, a battery optimization path planning system (100) according to one embodiment uses GA to determine the timing and path for the robot to visit a charging station, thereby optimizing energy management so that the energy required during the entire mission can be used efficiently.

[0056] A battery optimization path planning system (100) according to one embodiment uses GA to set an initial population of various possible paths. Each path set includes a charging station in a specific way. To evaluate the efficiency of each path, a fitness function is used to evaluate the excellence of each solution based on the total time and energy usage of the path. After the excellent solutions are selected, crossover and mutation occur among them to generate new solutions, and this process is repeated until an optimal solution is found that simultaneously optimizes the path and energy usage in the CPP process.

[0057] A battery optimization path planning system (100) according to one embodiment can operate to satisfy the conditions of [Equation 1] so that a robot can cover all designated areas within a specific environment without consuming the battery.

[0058]

[0059] Here, A represents the total area to be covered by the robot, and Q' represents the robot's path. That is, a battery optimization path planning system (100) according to one embodiment can establish a path plan to minimize the total movement and charging time T through path Q' as shown in [Equation 1]. At this time, each node n, which represents the points that the robot must visit i The sum of the areas covered using the sensor detection range Rs must be equal to A.

[0060] Nodes are points configured to allow the robot to effectively cover the entire area; each node must be within the robot's sensor detection range (Rs), and by passing through the point, the robot can complete tasks such as monitoring, inspection, or cleaning of the given area.

[0061] Nodes can be generated using specific algorithms; for example, a normal vector-based node generation method can be used to select orthogonal points between consecutive nodes (this allows for a design that satisfies sensor detection range conditions while minimizing the number of required nodes). One of the core objectives of CPP is to plan an optimal path by efficiently connecting these configured nodes.

[0062] At this time, each node n i Remaining battery Br(n at i)) must be greater than or equal to the minimum battery threshold Bmin. Under these conditions, it can be guaranteed that the robot does not stop due to low battery at any stage of path Q'. T(Q' is the total travel and charging time through path Q', and Coverage(n i , Rs) is node n for the sensor range of Rs. i It represents the area covered. The battery constraint is for each node n of path Q'. i Remaining battery Br(n i It is possible to prevent battery depletion by ensuring that ) is always greater than the minimum threshold Bmin.

[0063] In addition, when the robot is equipped with a sensor of a specific detection range Rs, the selection of nodes to visit by the robot considers the sensor range represented by [Equation 2].

[0064]

[0065] That is, all nodes n i , n j The distance between d(n i , n j ) must be within the sensor range Rs. To this end, the map generation unit (110) of the battery optimization path planning system (100) according to one embodiment generates a map containing the robot's charging station as shown in FIG. 2.

[0066] A map generation unit (110) according to one embodiment may generate a map using a simulated environment or Simultaneous Localization and Mapping (SLAM) technology. The generated map displays the locations of multiple charging stations (C=(c1,c2,...,cm)) as shown in FIG. 2. A position setting unit (120) of a battery optimization path planning system (100) according to one embodiment sets the initial position and target position of the robot on the generated map.

[0067] The full coverage environment map of Fig. 2 is filled with uniformly distributed Rs node points, which represent the nodes that the robot must visit, and the charging station points indicate the locations of battery charging stations. The robot in the full coverage environment must visit all Rs node points to cover the entire area. Rs represents the sensor range of the robot, and only nodes within this range can be detected and processed by the robot.

[0068] The limited coverage environment map of Fig. 2 represents a complex environment where only specific areas (coverage areas) need to be covered. In this environment, a path must be planned so that the robot does not pass through non-coverage areas. The full coverage environment of the generated map allows the robot to efficiently cover as many areas as possible, while the limited coverage environment restricts the robot to performing tasks only within specific areas. Such environmental settings are essential to optimize the robot's sensor and battery performance and to maximize operational efficiency under actual field conditions.

[0069] A graph generation unit (130) of a battery optimization path planning system (100) according to one embodiment constructs a graph that is the movement area of ​​a robot based on minimum coverage nodes. For example, a graph generation unit (130) according to one embodiment may construct a graph (G = (N, E)) using a sensor detection range Rs. N represents a node that the robot must cover, and E represents an edge that is an edge connecting these nodes. A path planning unit (140) according to one embodiment may perform CPP using ACO to satisfy the conditions shown in [Equation 3].

[0070]

[0071] The distance matrix E between nodes is configured to generate graph G, and Eij represents the distance between nodes i and j, and x ij is a binary variable indicating whether to move from node i to j (indicating whether node i and j are connected (1) or not (0)). In this case, there must be exactly one path starting from each node i (meaning every node must be connected to another node exactly once), and there must be exactly one path entering each node j (meaning every node must be connected to another node exactly once).

[0072] When the coverage start is determined by the initial position of the robot, the result of this optimized path can be sequentially expressed as Q = [q1, q2, q3, ..., qN] (where q1 is the starting node and qN is the last node). That is, the path planning unit (140) according to one embodiment determines X connecting each node i and j. ij This can be performed by using ACO to make probabilistic decisions based on the amount of pheromones left by the ant as it moves between nodes. In this way, the path planning unit (140) according to one embodiment finds the optimal connection between each node using ACO. The edge matrix E can be converted from a distance concept to a time concept to take into account both charging station visits and battery constraints, as shown in [Equation 4].

[0073]

[0074] Distance matrix E G represents the distance between a node and a charging station (each element of the matrix contains the distance from a specific node or charging station to another node or charging station). The travel time matrix T uses the average speed v to form the distance matrix E G It can be calculated in (the T matrix represents the time required for movement between each node and charging station). A path planning unit (140) according to one embodiment is E of [Equation 4] GBased on this, a CPP considering battery constraints can be planned as shown in [Equation 5] and [Equation 6].

[0075]

[0076] As shown in [Equation 5], the movement and charging time are minimized through the robot's path Q', T qi qi+1 is node q i from q i+1 Travel time to, T qi ci Wow T ci qi+1 is q i Nearest charging station C i q through i+1 Represents the time to travel to. a qiqi+1 wa b qiqi+1 is a binary decision variable that determines whether to move directly or go through a charging station. Each node pair q i wa q i+1 For , only one of the cases—direct travel or passing through a charging station—can be selected (a and b are binary variables representing mutually exclusive decisions). f is a dynamic and non-linear function. t (Br(L(q i ))) is a function that calculates the charging time that can be expressed as a complex number. L(q i ) is the length of the path.

[0077] A path planning unit (140) according to one embodiment can efficiently search and optimize by utilizing the evolutionary characteristics of GA to minimize movement and charging time through the robot's path Q' of [Equation 5]. That is, GA can be applied to find the optimal combination by generating various possible solutions, selecting the best solution among them, and generating a new solution through a crossover and mutation process.

[0078]

[0079] The robot is node q i Remaining battery Br(L(q) when following the path ati By ensuring that )) is greater than a specific threshold Bmin(i), it is possible to prevent the robot from failing to complete the mission due to a lack of energy. A battery optimization path planning system (100) according to one embodiment may use a normalized CPP to generate an optimal node sequence Q, and then introduce duplicate nodes between each pair of node sequences of Q. Each duplicate node may be a previous node or a charging station closest to the previous node. At this time, a path planning unit (140) according to one embodiment may determine the nodes using a genetic algorithm, which is a heuristic algorithm.

[0080] A battery optimization path planning system (100) according to one embodiment can form a graph based on a map including a charging station and a sensor detection range as shown in the framework for coverage path planning (CPP) integrating battery constraints of FIG. 3, and can perform heuristic path optimization, introduce node redundancy for energy management, and final path optimization using an energy-aware CPP solver.

[0081] A map generation unit (110) of a battery optimization path planning system (100) according to one embodiment generates a map of the area where a robot will be active as shown in FIG. 3, and displays the location of a charging station on this map (result of map generation including charging station). A graph generation unit (130) generates a graph G=(N,E) considering the sensor detection range Rs. The graph represents the path and connection structure that the robot can move along. At this time, the battery optimization path planning system (100) according to one embodiment selects a minimum coverage node that allows the entire area to be efficiently covered using only a minimum number of nodes. This is an initial step for path optimization, and is a process of selecting a node that can cover the entire area with the minimum necessary movement.

[0082] Subsequently, the path planning unit (140) solves the normal CPP problem using a heuristic approach. In this stage, various heuristic algorithms are applied to solve the complex path optimization problem. Next, the path is reconstructed considering energy constraints. This introduces additional nodes (e.g., charging stations or other strategic locations) for energy management to prevent energy shortage problems that may occur while the robot performs its mission. Next, the final path optimization is performed using an energy-aware CPP solver.

[0083] In this process, decisions regarding movement between nodes are made in a binary manner (this includes the process of determining whether the robot will move directly to the next node or visit a charging station for recharging). This framework provides a comprehensive approach to efficient and energy-efficient robot path planning. Considering battery constraints can significantly improve robot reliability and sustainability, particularly in large-scale or long-term automated operations.

[0084] A battery optimization path planning system (100) according to one embodiment reduces the total travel time by selecting the optimal charging time along the path through a hierarchical optimization process as described above. That is, it evaluates whether a charging station must be visited at each node along the determined coverage path and determines duplicate nodes. This decision process includes a binary choice of choosing between staying at the previous node (or moving to the next node) or visiting a charging station.

[0085] In this case, since the number of possible cases increases exponentially depending on the number of nodes N, the battery optimization path planning system (100) according to one embodiment applies a genetic algorithm, which is one of the heuristic approaches.

[0086] That is, the path planning unit (140) of the battery optimization path planning system (100) according to one embodiment creates an initial population of chromosomes by randomly assigning the number of charging station visits C, and evaluates the fitness of each chromosome based on the total charging and travel time. After creating the initial population in this way, selection, crossover, mutation, and replacement are performed sequentially to converge to an optimal solution over successive generations, thereby enabling the derivation of the optimal movement path of the robot.

[0087] In genetic algorithms, dominant chromosomes (e.g., paths with short travel times and low total charge times) are selected from a chromosome population (a collection of possible paths). The selection process can be performed by a fitness function, which evaluates the efficiency of each chromosome based on these criteria. The fitness function for chromosome k is defined to minimize total travel and charge times, as shown in [Equation 7].

[0088]

[0089] The function f(A(k)) is an objective function that enables the robot to select the optimal path by minimizing the total movement and charging times of path A(k), where A(k)=[aq1q2,aq2q3,...,aqN-1qN] is a vector representing the movement decisions between each pair of nodes.

[0090] The function f(A(k)) is node q i wa q i+1 Direct travel time between, charging station C i Time required to visit *, and related charging time f t (Br(L(q i Calculate the total cost by considering ))) variable a qiqi+1 (k) and b qiqi+1 (k) is a binary decision variable representing whether to consider a direct path or a charging station, respectively. qiqi+1 (k) and b qiqi+1The sum of (k) is always made to be 1 (so that only one of these options is selected for each step of the path).

[0091] Hereinafter, a path planning method for battery optimization according to an embodiment of the present invention is described based on the above description.

[0092]

[0093] FIG. 4 is a path planning method for battery optimization according to an embodiment of the present invention, and is an example of a path planning method for battery optimization using the path planning system for battery optimization of FIG. 1 of the present invention.

[0094] Referring to FIG. 4, a multimodal map construction method based on multi-object tracking according to an embodiment of the present invention generates a map including a robot's charging station by a map generation unit (110) (S410), and sets the initial position and target position of the robot by a position setting unit (120) (S420). After generating a graph representing the robot's movement area based on a minimum coverage node by a graph generation unit (130) (S430), a path planning unit (140) plans the robot's path based on a hierarchical heuristic algorithm (S440).

[0095] At this time, the path planning unit (140) can determine the initial path sequence of the robot based on an ant colony optimization algorithm. Specifically, the path planning unit (140) can determine the initial path sequence of the robot based on an ant colony optimization algorithm that adjusts the pheromone intensity based on the battery status of the robot.

[0096] Additionally, the path planning unit (140) can plan the robot's path by applying a genetic algorithm for introducing duplicate nodes and optimizing the final path to the determined initial path sequence. At this time, the path planning unit (140) can plan the robot's path by applying a genetic algorithm to the initial path sequence in which at least one of crossover and mutation of the genetic algorithm is performed based on the robot's battery status.

[0097] Afterwards, the battery charging unit (150) monitors the battery status of the robot as the robot moves along the planned robot path (S450), and based on the results of the monitoring, can arrive at a charging station and charge the battery (S460).

[0098]

[0099] As explained above, the present invention utilizes a hierarchical heuristic approach based on ant colony optimization and genetic algorithms to monitor battery status in real time and plan optimal charging times and paths, thereby enabling effective coverage of the entire area while minimizing the robot's battery consumption.

[0100]

[0101] The above embodiments may be implemented using various forms of computing means including one or more processors, memory, and storage means. Additionally, a network interface connected to a wired or wireless network may be included. The processor may be a central processing unit or a semiconductor device that executes processing instructions stored in memory and / or storage units. The memory and storage units may include volatile storage media or non-volatile storage media. For example, the memory may include ROM and RAM. Accordingly, embodiments of the present invention may be implemented as a method implemented by a computer or as a non-transient computer-readable medium having computer-executable instructions stored on said computer. In one embodiment of the present invention, when executed by a processor, the computer-readable instructions may perform a method according to at least one aspect of the present invention.

[0102] Although the present invention has been described with reference to illustrated embodiments as above, these are merely exemplary, and it will be obvious to those skilled in the art that various modifications, changes, and equivalent alternative embodiments are possible without departing from the gist and scope of the present invention. For example, the position setting unit (120) and the graph generation unit (130) may be implemented as a single integrated module or divided into two or more devices. Accordingly, the true technical scope of protection of the present invention should be determined by the technical concept of the appended claims.

[0103] The present invention can be used in systems that plan an optimal path for robots or autonomous vehicles to effectively move within a driving area.

Claims

1. A map generation unit that generates a map including a robot's charging station; A position setting unit for setting the initial position and target position of the above robot; A graph generation unit that generates a graph based on minimum coverage nodes; and It includes a path planning unit that plans the path of the robot based on a hierarchical heuristic algorithm, and The above graph represents the movement area of ​​the robot, a path planning system for battery optimization.

2. In Claim 1, The path planning unit above determines the initial path sequence of the robot based on an ant colony optimization algorithm, a path planning system for battery optimization.

3. In Claim 2, A path planning system for battery optimization, wherein the path planning unit determines the initial path sequence of the robot based on an ant colony optimization algorithm that adjusts pheromone intensity based on the battery status of the robot.

4. In Claim 1, A path planning system for battery optimization, wherein the path planning unit plans the path of the robot by introducing duplicate nodes into the determined initial path sequence and applying a genetic algorithm for final path optimization.

5. In Claim 4, A path planning system for battery optimization, wherein the path planning unit plans the path of the robot by applying a genetic algorithm to the initial path sequence, wherein at least one of the crossover and mutation of the genetic algorithm is performed based on the battery state of the robot.

6. In Claim 4, A path planning system for battery optimization, further comprising a battery charging execution unit that causes the robot to arrive at a charging station and charge the battery based on the monitored battery status of the robot as it moves along the planned path of the robot.

7. A step of generating a map containing robot charging stations; Step of setting the initial position and target position of the above robot; A step of generating a graph based on minimum coverage nodes; and It includes the step of planning the path of the robot based on a hierarchical heuristic algorithm, and The above graph represents the movement area of ​​the robot, a path planning method for battery optimization.

8. In Claim 7, The step of planning the robot's path based on the above hierarchical heuristic algorithm is A path planning method for battery optimization comprising the step of determining an initial path sequence of the robot based on an ant colony optimization algorithm.

9. In Claim 8, The step of determining the initial path sequence of the robot based on the above ant colony optimization algorithm A path planning method for battery optimization, comprising the step of determining an initial path sequence of the robot based on an ant colony optimization algorithm that adjusts pheromone intensity based on the battery status of the robot.

10. In claim 8, The step of planning the robot's path based on the above hierarchical heuristic algorithm is A step of determining an initial path sequence of the robot based on an ant colony optimization algorithm; and A path planning method for battery optimization, comprising the step of planning the path of the robot by introducing duplicate nodes into the determined initial path sequence and applying a genetic algorithm for final path optimization.

11. In Claim 10, The step of planning the path of the robot by applying a genetic algorithm for introducing duplicate nodes and optimizing the final path to the initial path sequence determined above is A path planning method for battery optimization comprising the step of planning a path of the robot by applying a genetic algorithm to the initial path sequence, wherein at least one of crossover and mutation of the genetic algorithm is performed based on the battery state of the robot.

12. In Claim 10, The step of planning the path of the robot by applying a genetic algorithm for introducing duplicate nodes and optimizing the final path to the initial path sequence determined above is A step of monitoring the battery status of the robot as the robot moves along the planned path of the robot; and A method for planning a path for battery optimization, comprising the step of arriving at the charging station and charging the battery based on the results of the above monitoring.

Images