Unmanned aerial vehicle dynamic path planning and obstacle avoidance method based on real-time environment perception

Abstract

The invention belongs to the technical field of unmanned aerial vehicle navigation and path planning, and provides an unmanned aerial vehicle dynamic path planning and obstacle avoidance method based on real-time environment perception, which comprises the following steps: judging whether the current flight path of an unmanned aerial vehicle needs to be re-planned according to real-time environment data of the current flight path of the unmanned aerial vehicle; if the current flight path of the unmanned aerial vehicle needs to be re-planned, generating a plurality of alternative paths according to the current position of the unmanned aerial vehicle; obtaining path feature data of the alternative path, and generating an alternative path collision risk value; generating an alternative path comprehensive evaluation value according to the estimated travel time, the estimated energy consumption, the airspace rule data and the alternative path collision risk value of the alternative path; generating an optimal alternative path according to the alternative path comprehensive evaluation value; according to the optimal alternative path, planning and obstacle avoidance are carried out on the unmanned aerial vehicle path; the method has the advantages that the dynamic path planning and obstacle avoidance of the unmanned aerial vehicle can be realized, and the flight safety, task efficiency and regulation compliance of the unmanned aerial vehicle are improved.

Description

Technical Field

[0001] This invention belongs to the field of UAV navigation and path planning technology, and particularly relates to a method for dynamic path planning and obstacle avoidance of UAVs based on real-time environmental perception. Background Technology

[0002] With the widespread application of drones in logistics, aerial surveying, and emergency rescue, their operating environments are becoming increasingly complex and dynamic, placing higher demands on the real-time performance, safety, and adaptability of flight paths. Especially in dynamic environments such as cities, forests, and mountains, drones need the ability to perceive environmental changes in real time and autonomously adjust their paths to achieve efficient and safe flight and obstacle avoidance. Therefore, researching a method for dynamic path planning and obstacle avoidance based on real-time environmental perception is of great significance.

[0003] Currently, UAV path planning and obstacle avoidance technologies mainly include static planning based on preset waypoints, local obstacle avoidance based on sensors (such as LiDAR and visual cameras), and some dynamic replanning methods based on environmental models. Existing methods mostly focus on path adjustment under a single obstacle avoidance trigger or fixed rules, such as executing preset obstacle avoidance actions after detecting an obstacle, or relying on historical map data for path optimization. Some advanced systems have introduced real-time environmental data, but they are still relatively limited in terms of path evaluation dimensions.

[0004] However, existing technologies have significant shortcomings: First, most methods are slow to respond to dynamic environmental changes and lack a quantitative assessment mechanism for the continuous safety of the path; second, they often only consider obstacle avoidance as a single factor and do not incorporate multi-dimensional environmental characteristics such as path width, vegetation density, and frequency of biological activity into the comprehensive assessment of collision risk; third, they do not fully integrate multiple objectives such as travel time, energy efficiency, and airspace compliance for coordinated optimization, resulting in a difficulty in balancing safety, efficiency, and compliance in the planning results. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a dynamic path planning and obstacle avoidance method for unmanned aerial vehicles (UAVs) based on real-time environmental perception, thus solving the aforementioned problems.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a dynamic path planning and obstacle avoidance method for unmanned aerial vehicles (UAVs) based on real-time environmental perception, specifically comprising: Obtain real-time environmental data of the drone's current flight path, and determine whether the drone's current flight path needs to be replanned based on the real-time environmental data of the drone's current flight path; If the current flight path of the drone needs to be replanned, several alternative paths will be generated based on the current location of the drone. Obtain path feature data of alternative paths, and generate collision risk values ​​for alternative paths based on the feature data of alternative paths; where path feature data refers to feature data in the flight path that may cause the drone to collide. Obtain estimated travel time, estimated energy consumption, and airspace rule data for alternative routes; A comprehensive evaluation model for alternative routes is established based on the estimated travel time, estimated energy consumption, airspace rule data, and collision risk values ​​of alternative routes, and a comprehensive evaluation value for alternative routes is generated. The optimal alternative path is generated based on the comprehensive evaluation of the alternative paths. Based on the optimal alternative path, the drone's path is planned and obstacle avoidance is performed.

[0007] Based on the above technical solutions, the present invention also provides the following optional technical solutions: Further technical solutions: The method for determining whether the current flight path of the drone needs to be replanned specifically includes: Based on real-time environmental data of the current flight path of the drone, a safety factor for the current flight path is generated; Based on the safety factor of the current flight path, determine whether the current flight path of the drone needs to be replanned.

[0008] Further technical solution: The method for generating the safety factor of the current flight path specifically includes: Through the formula: ; Generate the safety factor of the current flight path ; In the formula, This represents the normalized value of the i-th real-time environmental data point along the current flight path of the drone. This represents the weighting coefficient of the i-th real-time environmental data.

[0009] Further technical solution: The specific method for generating the collision risk value of the alternative path includes: Through the formula: ; Generate collision risk values ​​for alternative paths ; In the formula, F represents the frequency of historical environmental mutations in the candidate paths. Minimum frequency of historical environmental mutations in alternative paths This represents the maximum frequency of historical environmental mutations among the alternative paths. This represents the normalized value of the feature data of the j-th path among the candidate paths. This represents the weight coefficient of the j-th path feature among the candidate paths. , All are weighting ratio coefficients, and .

[0010] Further technical solution: The method for generating the comprehensive evaluation value of the alternative paths specifically includes: Based on the estimated travel time of the alternative routes, a timeliness loss factor is generated; Based on the estimated energy consumption of alternative routes, a range impact factor is generated; Generate a legality assessment factor based on the airspace rule data of the alternative paths; A comprehensive evaluation model for alternative routes is established based on timeliness loss factors, range impact factors, legality assessment factors, and alternative route collision risk values, generating a comprehensive evaluation value for alternative routes.

[0011] Further technical solution: The specific method for generating the timeliness loss factor includes: Through the formula: ; Generate timely loss factor ; In the formula, This indicates the estimated travel time for the alternative routes. This indicates the estimated travel time of the drone's current flight path.

[0012] Further technical solution: The specific method for generating the range influencing factor includes: Through the formula: ; Generate battery life influencing factors ; In the formula, This indicates the estimated energy consumption of the alternative routes. This indicates the estimated energy consumption of the drone's current flight path.

[0013] Further technical solution: The specific method for generating the legality assessment factor includes: Through the formula: ; Generate a legality assessment factor L; In the formula, This represents the k-th spatial rule data of the candidate paths. For conditional flag functions, This represents the weight coefficient of the k-th spatial rule.

[0014] Further technical solution: The expression of the comprehensive evaluation model for alternative paths is: ; In the expression, Score represents the overall evaluation value of the alternative paths. This represents the collision risk value of the alternative paths. This represents the timeliness loss factor. This represents the range-affecting factor, and L represents the legality assessment factor. , , All are weighting ratio coefficients, and .

[0015] Further technical solution: The method for generating the optimal alternative path specifically includes: Through the formula: ; Generate the optimal alternative path ; In the formula, This represents the g-th alternative path. This represents the comprehensive evaluation value of the candidate path for the g-th candidate path. This represents the set of all alternative paths.

[0016] This invention provides a method for controlling the cooling fan of a desktop computer, which has the following advantages compared with the prior art: This invention acquires real-time environmental data and determines whether replanning is needed, generates alternative paths, calculates collision risk values, obtains estimated travel time, estimated energy consumption, and airspace rule data, establishes a comprehensive evaluation model to generate evaluation values, and selects the optimal path accordingly. This achieves dynamic path planning and obstacle avoidance, enabling real-time response to environmental changes. By comprehensively evaluating multiple factors such as collision risk, travel time, energy efficiency, and airspace compliance, it generates the optimal path, thereby improving the flight safety, mission efficiency, and regulatory compliance of UAVs. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a dynamic path planning and obstacle avoidance method for unmanned aerial vehicles (UAVs) based on real-time environmental perception, provided by the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0019] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.

[0020] Please see Figure 1 The present invention provides a method for dynamic path planning and obstacle avoidance of unmanned aerial vehicles (UAVs) based on real-time environmental perception, comprising the following steps: Step S10: Obtain real-time environmental data of the current flight path of the drone, and determine whether the current flight path of the drone needs to be replanned based on the real-time environmental data of the current flight path of the drone. Step S20: If the current flight path of the UAV needs to be replanned, generate several alternative paths based on the current position of the UAV; The alternative paths include the drone's current flight path; this is to prevent other paths from being worse than the current flight path. Step S30: Obtain the path feature data of the alternative paths, and generate the collision risk value of the alternative paths based on the feature data of the alternative paths; Among them, path feature data refers to the feature data that may cause the drone to collide with the flight path; such as path width, density of branches in the path, frequency of appearance of flying creatures in the path, etc. Step S40: Obtain the estimated travel time, estimated energy consumption, and airspace rule data for the alternative routes; Step S50: Establish a comprehensive evaluation model for alternative routes based on the estimated travel time, estimated energy consumption, airspace rule data, and collision risk values ​​of alternative routes, and generate a comprehensive evaluation value for alternative routes. Step S60: Generate the optimal alternative path based on the comprehensive evaluation value of the alternative paths; Step S70: Plan and avoid obstacles for the drone based on the optimal alternative path; In step S10, real-time environmental data can be acquired in various ways. For example, the drone can be equipped with a visual sensor to identify obstacles or environmental changes on the path through image processing technology; or, the drone can be equipped with a lidar to construct a three-dimensional model of the surrounding environment through point cloud data, thereby perceiving the distance and shape of obstacles; in addition, whether replanning is needed can be determined based on a preset threshold.

[0021] In step S20, the method for generating alternative paths can be varied. For example, sampling-based path planning algorithms, such as the Quick Search Random Tree (RRT) algorithm or the Probabilistic Route Map (PRM) algorithm, can be used to randomly generate multiple feasible paths between the UAV's current position and the target point. Furthermore, the alternative paths include the UAV's current flight path. Including the current flight path as one of the alternative paths ensures that if no better alternative is found, the UAV can still choose to continue flying along the current path, avoiding a deterioration of the situation due to blindly switching paths.

[0022] In step S30, the path feature data can include multiple dimensions. For example, the path width can be analyzed using image recognition technology, or the density of branches in the path can be evaluated using a deep learning model. Furthermore, the frequency of flying creatures (e.g., birds) appearing in the path can be statistically analyzed using acoustic sensors or visual recognition technology. This path feature data can be quantified; for example, the path width can be normalized to a value between 0 and 1, and the density of branches can be evaluated based on pixel density.

[0023] In step S50, the comprehensive evaluation model can be a multi-objective optimization model designed to balance safety, efficiency, and compliance. For example, a linear weighted model can be used, assigning different weights to factors such as collision risk value, travel time, energy consumption, and airspace compliance, and then performing a weighted summation. Through this model, each alternative path is assigned a quantified comprehensive evaluation value, thereby reflecting the overall merits of the path.

[0024] In step S60, generating the optimal candidate path is typically accomplished by comparing the comprehensive evaluation values ​​of all candidate paths. A higher comprehensive evaluation value indicates a better path, and the path with the highest comprehensive evaluation value is selected as the optimal candidate path. In some cases, a minimum evaluation value threshold can also be set, and only paths that reach or exceed this threshold will be considered, with the optimal one selected from among them.

[0025] Finally, in step S70, once the optimal alternative path is determined, the UAV flight control system will receive the path information and use it as a new flight command. The UAV will fly according to the optimal path and continuously perform real-time environmental perception to deal with any emergencies that may occur during the flight.

[0026] Preferably, the present invention further proposes a method for determining whether the current flight path of the UAV needs to be replanned, specifically including: Step S11: Generate the safety factor of the current flight path based on the real-time environmental data of the current flight path of the UAV; Step S12: Based on the safety factor of the current flight path, determine whether the current flight path of the UAV needs to be replanned; In step S11, the current flight path safety factor is a quantitative indicator used to comprehensively assess the safety of the UAV's current flight path. It is generated based on real-time environmental data, aiming to transform complex environmental information into a single, comparable value for subsequent decision-making.

[0027] In step S12, this application uses the generated quantified safety coefficient as the decision-making basis, avoiding subjective judgment and improving the accuracy and automation of the judgment. For example, the judgment method can be based on setting a preset safety coefficient threshold. When the safety coefficient of the generated current flight path is lower than the threshold, it is determined that replanning is required; conversely, if it is higher than or equal to the threshold, the current path is considered safe and no replanning is required. In addition, a dynamic threshold adjustment strategy can also be adopted, which adjusts the judgment threshold of the safety coefficient in real time according to factors such as the drone's mission type, remaining battery power, and flight stage to adapt to different flight scenario requirements.

[0028] This application's solution addresses the ambiguity in determining whether a UAV's current flight path needs replanning by introducing a quantitative evaluation mechanism. Specifically, after acquiring real-time environmental data of the UAV's current flight path, instead of making subjective judgments, it first generates a current flight path safety coefficient based on this data, comprehensively reflecting the safety status of the current path. This safety coefficient quantifies the impact of various environmental factors (such as obstacle distribution and weather conditions) into a unified value. Subsequently, the system compares this generated safety coefficient with a preset safety threshold. If the safety coefficient is lower than the threshold, it indicates a high safety risk, triggering a path replanning process; conversely, if the safety coefficient reaches or exceeds the threshold, the current path is considered safe, and replanning is unnecessary. This approach transforms the judgment process into automated decision-making based on quantitative indicators, enabling the UAV to more accurately and promptly identify potential risks and make corresponding path adjustments, thereby effectively improving the safety and decision-making efficiency of UAV flights.

[0029] Through the above technical solution, this application provides a clear and quantifiable judgment mechanism for determining whether the current flight path of a UAV needs to be replanned. By generating a safety factor for the current flight path based on real-time environmental data and using this factor as the judgment criterion, it avoids the fuzzy judgments or subjective decisions that may exist in traditional methods. This judgment method based on a quantified safety factor enables the UAV to more accurately assess the risk level of the current flight environment, thereby initiating path replanning in a timely manner when necessary, effectively avoiding potential dangers and ensuring flight safety. At the same time, when the path safety factor is within an acceptable range, the system will not perform unnecessary replanning, thereby reducing the consumption of computing resources and improving the efficiency of path planning and the stability of the system.

[0030] Preferably, the present invention further proposes a method for generating the safety factor of the current flight path, specifically including: Through the formula: ; Generate the safety factor of the current flight path ; In the formula, This represents the normalized value of the i-th real-time environmental data point along the current flight path of the drone. This represents the weight coefficient of the i-th real-time environmental data. Among them, the normalized value of the i-th real-time environmental data in the current flight path of the UAV Its purpose is to unify real-time environmental data (such as wind speed, obstacle density, visibility, etc.) with different dimensions and ranges onto a standardized scale, so as to perform effective weighted summation calculations. Normalization eliminates the differences in dimensions between data, ensuring the fairness and comparability of various environmental data in the calculation of safety factors; The weighting coefficient of the i-th real-time environmental data This reflects the importance or influence of real-time environmental data in assessing the safety factor of the current flight path. By assigning different weights to different environmental factors, key factors with a greater impact on flight safety can be highlighted, making the safety factor calculation results more targeted and accurate. For example, subjective values ​​can be assigned based on expert experience or domain knowledge; when flying in mountainous areas, the weight of terrain complexity may be higher than that in plains.

[0031] Through the above technical solution, this application provides a quantitative and objective method for generating the safety factor of the current flight path. By normalizing multiple real-time environmental data points of the UAV's current flight path and assigning corresponding weight coefficients for weighted summation, the overall safety status of the current flight path can be comprehensively assessed. This avoids the uncertainty caused by subjective judgment, making the assessment of flight risks more accurate and scientific. Path replanning based on this safety factor can more timely and effectively identify potential hazards and trigger obstacle avoidance mechanisms, thereby significantly improving the flight safety, decision-making efficiency, and mission reliability of UAVs in complex environments.

[0032] Preferably, the present invention further proposes a method for generating the collision risk value of the alternative path, specifically including: Through the formula: ; Generate collision risk values ​​for alternative paths ; In the formula, F represents the frequency of historical environmental mutations in the candidate paths. Minimum frequency of historical environmental mutations in alternative paths This represents the maximum frequency of historical environmental mutations among the alternative paths. This represents the normalized value of the feature data of the j-th path among the candidate paths. This represents the weight coefficient of the j-th path feature among the candidate paths. , All are weighting ratio coefficients, and ; The historical environmental abruptness frequency of the alternative route refers to the frequency of environmental abruptness events (such as the sudden appearance of obstacles or abrupt weather changes) that have occurred along the alternative route in the past. This reflects the inherent instability or danger of the route. It can be obtained by analyzing historical flight data, utilizing dynamic obstacle records in Geographic Information System (GIS) data, or performing pattern recognition and prediction on historical environmental data through machine learning models; The minimum and maximum historical environmental mutation frequencies of the alternative paths are used to normalize the historical environmental mutation frequencies, mapping them to a standardized range for weighted calculation with other risk factors. The weight coefficients of the j-th path feature among the candidate paths are used to measure the importance of different path features in terms of their impact on collision risk. For example, path width may have a higher weight than the frequency of flying creature occurrences. These weight coefficients can be set through expert experience, historical data regression analysis, or trained and optimized using machine learning algorithms (such as decision trees and neural networks). Weighting ratio coefficient , This is used to balance the relative importance of historical environmental change frequency and current path characteristic data in the overall collision risk assessment. These factors ensure that the contribution of each risk source can be flexibly adjusted according to the actual application scenario, and that the total weight is 1, maintaining the reasonableness of the assessment results. For example, in areas with rapid environmental changes, the weighting coefficients are adjusted accordingly. It can be set higher; in areas with relatively stable environments, the weighting ratio coefficient... It can be set higher.

[0033] Specifically, this method first normalizes the frequency of historical environmental mutations, transforming it into a risk component reflecting the historical uncertainty of the path. Simultaneously, it normalizes the path characteristic data of the candidate paths and, combined with their corresponding weight coefficients, calculates a risk component reflecting the current environmental characteristics. Then, it uses weighting ratio coefficients... , The two risk components are weighted and summed to obtain the final collision risk value for the alternative paths. This assessment mechanism, which combines historical experience with real-time perception, makes the assessment of collision risk more comprehensive and accurate.

[0034] Through the above technical solution, this application effectively addresses the problem that relying solely on real-time path feature data is insufficient for assessing collision risk. By introducing the historical environmental mutation frequency F of alternative paths and weighting it with the real-time acquired path feature data, the generation of collision risk values ​​for alternative paths becomes more comprehensive and accurate. This comprehensive assessment method not only considers the direct impact of the current environment but also incorporates potential, less noticeable risk factors from historical experience, thereby significantly improving the accuracy and predictability of collision risk assessment. Therefore, in the process of UAV dynamic path planning, high-risk paths can be identified and avoided more effectively, reducing the probability of UAV collisions and thus improving the safety, reliability, and mission completion efficiency of UAV flights.

[0035] Preferably, the present invention further proposes a method for generating the comprehensive evaluation value of the alternative paths, specifically including: Step S51: Generate a timeliness loss factor based on the estimated travel time of the alternative routes; Step S52: Generate range impact factors based on the estimated energy consumption of the alternative routes; Step S53: Generate a legality assessment factor based on the airspace rule data of the candidate paths; Step S54: Establish a comprehensive evaluation model for alternative routes based on the timeliness loss factor, range impact factor, legality assessment factor, and alternative route collision risk value, and generate a comprehensive evaluation value for alternative routes; Among them, the timeliness loss factor is a quantitative indicator that measures the deviation or loss of alternative paths in terms of time efficiency relative to a specific benchmark (such as the current flight path or ideal flight time). The endurance impact factor is a quantitative indicator that assesses the impact of alternative routes on the energy consumption of drones, reflecting the potential impact of route selection on the endurance or energy efficiency of drones. The legality assessment factor is a quantitative indicator that measures whether alternative routes comply with relevant airspace flight regulations and restrictions, ensuring the compliance of drone flight activities.

[0036] The proposed solution, when generating a comprehensive evaluation value for alternative routes, first refines the evaluation process into multiple independent evaluation dimensions. Specifically, for the estimated travel time of alternative routes, a timeliness loss factor is generated in step S51 to quantify the impact of time efficiency; for the estimated energy consumption of alternative routes, a range impact factor is generated in step S52 to evaluate the efficiency of energy consumption; and for the airspace rule data of alternative routes, a legality evaluation factor is generated in step S53 to ensure flight compliance. These factors, along with the obtained collision risk values ​​of alternative routes, are integrated into a comprehensive evaluation model for alternative routes in step S54. This model systematically considers multiple key factors such as time, energy consumption, compliance, and safety, quantifies them, and performs comprehensive calculations to generate a comprehensive evaluation value for alternative routes that fully reflects the overall merits and demerits of the alternative routes. This step-by-step quantification and comprehensive evaluation mechanism makes the evaluation of alternative routes no longer a single-dimensional consideration, but a multi-dimensional and systematic decision-making process, effectively solving the problem of how to effectively integrate heterogeneous data to comprehensively evaluate the merits and demerits of routes.

[0037] Through the aforementioned technical solution, this application enables a systematic and quantitative evaluation of multiple key factors involved in UAV path planning, including estimated travel time, estimated energy consumption, airspace rule data, and collision risk values ​​of alternative paths. By introducing timeliness loss factors, endurance impact factors, and legality assessment factors, and combining them with collision risk values ​​of alternative paths, a multi-dimensional comprehensive evaluation model for alternative paths is constructed. This allows UAVs to consider not only safety but also flight efficiency, energy consumption, and regulatory compliance when selecting paths, thereby generating a more comprehensive, reasonable, and practically compliant comprehensive evaluation value for alternative paths. This refined evaluation mechanism significantly improves the intelligence level of path planning and the accuracy of decision-making, effectively avoiding potential risks or inefficiencies caused by insufficient consideration of a single factor.

[0038] Preferably, the present invention further proposes a method for generating the timeliness loss factor, specifically including: Through the formula: ; Generate timely loss factor ; In the formula, This indicates the estimated travel time for the alternative routes. This indicates the estimated travel time of the drone's current flight path; The timeliness loss factor is a dimensionless numerical value used to quantify the degree of change in estimated travel time between alternative routes and the current flight path of the UAV. This factor can intuitively reflect the increase or decrease in time cost that may result from choosing an alternative route, and is a key indicator for evaluating the time efficiency of a route.

[0039] The formula for calculating the timeliness loss factor is derived by comparing the relative difference between the estimated travel time of the alternative path and the estimated travel time of the UAV's current flight path. This calculation method can standardize the time loss of different paths, making them comparable in the comprehensive evaluation model.

[0040] The estimated travel time for an alternative path refers to the predicted time required for a UAV to reach its target point from its current location when flying along an alternative path. This time can be obtained in various ways, such as through simulation calculations based on factors like path length, the UAV's current speed, wind speed and direction along the path, terrain undulations, and potential speed-restricted areas along the path.

[0041] The estimated travel time of a drone's current flight path refers to the predicted time required for the drone to reach its target point from its current location if it continues to fly along its current flight path. This time can also be obtained using similar methods, such as dynamic estimation based on the remaining length of the current path, the drone's current speed, and real-time environmental data (such as wind speed); or calculation using pre-planned path information and drone performance parameters.

[0042] Through the above technical solution, this application provides a standardized and quantitative method for calculating the timeliness loss factor. This method can accurately measure the change in time efficiency of alternative paths relative to the current flight path, avoiding errors caused by subjective judgment or fuzzy evaluation. Therefore, introducing the timeliness loss factor calculated in this way into the comprehensive evaluation model of alternative paths can significantly improve the accuracy and reliability of the comprehensive evaluation results. This allows UAVs to more scientifically and rationally weigh time costs during dynamic path planning, ultimately selecting the truly optimal flight path and effectively improving the efficiency of UAV mission execution and the accuracy of decision-making.

[0043] Preferably, the present invention further proposes a method for generating the range influencing factor, specifically including: Through the formula: ; Generate battery life influencing factors ; In the formula, This indicates the estimated energy consumption of the alternative routes. This indicates the estimated energy consumption of the drone's current flight path; The endurance impact factor is a quantitative indicator used to measure the degree of change in energy consumption of alternative routes relative to the current flight path, thus reflecting the impact of choosing a new route on the drone's endurance. This factor can intuitively indicate whether the energy consumption of alternative routes increases or decreases, and the proportion of increase or decrease, providing a standardized measure of energy consumption impact for the comprehensive evaluation model.

[0044] The estimated energy consumption of the alternative paths in the formula can be obtained in several ways. For example, based on parameters such as the drone's aerodynamic characteristics, weight, speed, climb / descent angle, wind speed, and air density, combined with the path's geometric features (length, altitude changes), the energy consumption along the path can be calculated using physical equations. Alternatively, a machine learning model can be trained using historical flight data (including path, environmental conditions, and actual energy consumption), inputting the features of the alternative paths and environmental data to predict their energy consumption. Similarly, the estimated energy consumption of the drone's current flight path is obtained in a similar way to the estimated energy consumption of the alternative paths, but it applies to the path currently being executed.

[0045] Specifically, the system obtains the estimated energy consumption of each alternative path and the estimated energy consumption of the current flight path. Then, it uses the aforementioned formula to calculate the endurance impact factor for each alternative path. In this way, changes in energy consumption are standardized and incorporated into the overall evaluation framework, ensuring that the comprehensive evaluation value of the alternative paths fully reflects their energy efficiency. This ensures that when generating the optimal alternative path, factors such as timeliness, legality, and collision risk are considered, while also fully weighing the impact of energy consumption on the drone's endurance, thus avoiding the selection of paths that might cause the drone to run out of power prematurely or fail to complete the mission.

[0046] Through the above technical solution, this application provides a clear and quantifiable method for calculating the endurance impact factor, effectively solving the problem of unclear energy consumption considerations in path evaluation. This enables the comprehensive evaluation model of alternative paths to more accurately reflect the actual impact of different paths on the UAV's endurance, thus allowing for a more scientific and reasonable balance between flight time, safety, legitimacy, and energy consumption during dynamic path planning. Ultimately, this helps the UAV select the optimal flight path that effectively avoids obstacles, meets mission requirements, and maximizes endurance or optimizes energy use, significantly improving the reliability and efficiency of UAV missions.

[0047] Preferably, the present invention further proposes a method for generating the legality assessment factor, specifically including: Through the formula: ; Generate a legality assessment factor L; In the formula, This represents the k-th spatial rule data of the candidate paths. For conditional flag functions, This represents the weight coefficient of the k-th spatial rule; The legality assessment factor is generated in a way that provides an indicator to quantify the degree of compliance of alternative routes with airspace rules. This generation method combines the compliance assessment results of multiple airspace rule data through a weighted summation method, thereby providing a unified value that reflects the overall performance of alternative routes in complying with airspace rules. Specifically, this application multiplies and sums the evaluation results of each airspace rule (given by the conditional flag function) with their corresponding weight coefficients, thereby providing a structured and quantifiable method to calculate the legality evaluation factor and ensure the objectivity and consistency of the evaluation process. The k-th airspace rule data for the candidate path is specific airspace rule information related to the candidate path. For example, the k-th airspace rule data can be numerical data such as the difference between the drone's flight altitude and the upper limit of the prescribed altitude, the minimum distance between the drone and the boundary of the no-fly zone, or the minimum distance between the drone and the boundary of a residential area; it can also be Boolean data indicating whether a specific airspace (such as a temporary control zone or a military exercise zone) has been entered. This data serves as input to a conditional flag function to determine whether the candidate path conforms to specific airspace rules.

[0048] The conditional flag function determines the compliance of the input spatial rule data. For example, it can output 1 when the k-th spatial rule data satisfies the spatial rule, and output 0 otherwise. Alternatively, it can output a positive value (indicating compliance level) when the k-th spatial rule data satisfies the spatial rule, and output a negative value or 0 (indicating non-compliance or violation level) when the k-th spatial rule data does not satisfy the spatial rule. The weight coefficient of the k-th airspace rule reflects the importance of different airspace rules in the legality assessment. These weight coefficients can be set by expert experience based on factors such as the strictness of the airspace rules, their impact on flight safety, and the mandatory nature of laws and regulations; or they can be automatically learned by machine learning algorithms trained and optimized based on historical flight data and violation records. By introducing weight coefficients, the legality assessment can distinguish the importance of different airspace rules and more accurately reflect the overall compliance of alternative routes.

[0049] This application's solution identifies all relevant airspace rule data for alternative paths and applies a conditional flag function to each data point for compliance assessment, transforming the raw rule data into a quantified compliance score. Subsequently, the compliance score for each rule is multiplied by its corresponding weight coefficient to reflect the varying importance of different airspace rules. For example, the weight of violating a no-fly zone rule may be significantly higher than the weight of slightly exceeding the recommended flight altitude. Finally, all weighted compliance scores are summed to obtain a legality assessment factor. This structured weighted summation method ensures that the legality assessment factor comprehensively, objectively, and precisely measures the compliance of alternative paths, thus solving the problem of effectively quantifying the degree of compliance of alternative paths with complex airspace rules with different weights. This legality assessment factor is then integrated into the comprehensive evaluation model of alternative paths, working in conjunction with timeliness loss factors, endurance impact factors, and alternative path collision risk values. This allows UAVs to consider not only flight efficiency and safety but also full compliance with airspace regulations during dynamic path planning, thereby generating optimal alternative paths that are both efficient and compliant.

[0050] By employing the aforementioned technical solution and generating a weighted summation method to generate legality assessment factors, this application can comprehensively and quantitatively evaluate the degree of compliance of alternative paths with various airspace rules. This method not only considers the differences in importance among different airspace rules, distinguishing them through weight coefficients, but also transforms complex rule judgments into unified numerical values ​​through conditional flag functions, thereby avoiding biases from subjective judgments. This enables UAVs to more accurately identify and select paths that comply with airspace regulations during dynamic path planning, significantly improving the compliance and safety of flight missions and effectively avoiding risks and penalties that may result from violations of airspace regulations. Compared to relying solely on experience or simple judgment, this solution provides a more scientific and reliable basis for legality assessment of UAV path planning.

[0051] Preferably, the present invention further proposes the following expression for the comprehensive evaluation model of the alternative paths: ; In the expression, Score represents the overall evaluation value of the alternative paths. This represents the collision risk value of the alternative paths. This represents the timeliness loss factor. This represents the range-affecting factor, and L represents the legality assessment factor. , , All are weighting ratio coefficients, and ; Among them, the comprehensive evaluation value of the alternative path is a quantitative evaluation of the overall quality of the alternative path; the larger the value, the better the path. Weighting ratio coefficient , , These coefficients are used to adjust the relative importance of timeliness loss factors, endurance impact factors, and legitimacy assessment factors in the overall evaluation. These weighting coefficients can be preset or dynamically adjusted according to specific mission requirements, UAV type, or flight environment. For example, in missions with strict time requirements, they can be appropriately increased. The value; in tasks requiring high battery endurance, it can be improved. The value; in areas with strict airspace control, it can be increased. The sum of these weighting coefficients is usually set to 1 to ensure the proportionality of each factor's contribution.

[0052] This application's solution introduces a clear comprehensive evaluation model expression for alternative paths, considering the collision risk value of each alternative path as a multiplicative factor. This design significantly reduces the final comprehensive evaluation value of any alternative path with a high collision risk, thus prioritizing the elimination of unsafe paths during path selection and ensuring the paramount importance of flight safety. Furthermore, the model integrates timeliness loss factors, range impact factors, and legality evaluation factors through a weighted summation method. Specifically, and Factors originally representing "losses" or "negative impacts" are transformed into terms representing "benefits" or "positive impacts," with higher values ​​indicating better performance. The legality assessment factor L directly reflects the compliance of the path. This is achieved through weighted proportional coefficients. , , By weighting these three factors, the importance of time efficiency, energy efficiency, and compliance in the overall assessment can be flexibly adjusted according to the actual application scenario and task priority. For example, in emergency rescue missions, timeliness can be given higher weight; in long-distance inspection missions, endurance can be given higher weight. This multiplicative and additive assessment model not only comprehensively and objectively quantifies the various performance indicators of alternative paths, but also ensures the absolute priority of safety through multiplicative processing of collision risk, while achieving flexibility and adaptability in the assessment through adjustable weight coefficients. The final comprehensive assessment value of the alternative paths accurately reflects the overall advantages and disadvantages of each alternative path, providing a scientific and reliable decision-making basis for UAV dynamic path planning.

[0053] The above technical solution provides a clear and quantifiable comprehensive evaluation model expression for alternative paths, resolving issues such as unreasonable weight allocation and inaccurate evaluation results that may arise when integrating multi-dimensional evaluation factors. This model uses the collision risk value of alternative paths as a multiplicative factor, ensuring the priority of flight safety in path selection and effectively avoiding the neglect of safety risks in pursuit of other performance indicators. Simultaneously, by weighting and summing the timeliness loss factor, endurance impact factor, and legality evaluation factor, and introducing weight ratio coefficients, the evaluation process can flexibly adapt to different mission requirements and operating environments, achieving a balance between time efficiency, energy efficiency, and compliance. Ultimately, the model can generate a comprehensive, objective, and comparable comprehensive evaluation value for alternative paths, thereby significantly improving the decision-making quality and reliability of UAV dynamic path planning and ensuring the safe, efficient, and compliant operation of UAVs in complex environments.

[0054] Preferably, the present invention further proposes a method for generating the optimal alternative path, specifically including: Through the formula: ; Generate the optimal alternative path ; In the formula, This represents the g-th alternative path. This represents the comprehensive evaluation value of the candidate path for the g-th candidate path. It represents the set of all alternative paths; The formula for calculating the optimal alternative path defines the criterion for determining the optimal alternative path; it represents the selection of the optimal alternative path from the set of all alternative paths. In the process, the comprehensive evaluation value of the alternative paths is selected. Alternative paths corresponding to reaching the maximum value This selection method ensures that the chosen path performs optimally overall after comprehensively considering multiple dimensions such as safety, timeliness, energy consumption, and legality. All alternative paths This refers to the set of all available alternative paths generated for a drone. This set includes all potential, feasible flight paths, including the drone's current flight path.

[0055] In the dynamic path planning and obstacle avoidance process of UAVs, when the system determines that path replanning is necessary, it generates a series of alternative paths. To select the most suitable path for the UAV from these alternatives, this application introduces an optimal path selection mechanism based on a comprehensive evaluation value. Specifically, for the set... Each alternative path in The system will calculate the comprehensive evaluation value of the corresponding alternative paths. By using the formula for calculating the optimal alternative path, the system can accurately identify the path with the highest comprehensive evaluation value among all alternative paths and determine it as the optimal alternative path. This mechanism ensures that the selected path not only considers obstacle avoidance safety but also flight efficiency, economy, and legality, thus providing a comprehensively optimized flight plan for drones in complex and ever-changing environments. Compared to traditional methods that only consider a single factor (such as the shortest path or lowest risk), this solution significantly improves the intelligence and robustness of path planning by comprehensively and quantitatively evaluating multi-dimensional indicators and making decisions by maximizing the evaluation value. This enables drones to make more informed and safer flight decisions in dynamic environments.

[0056] Through the above technical solution, this application ensures that the selected optimal alternative path during the UAV dynamic path planning process is the best choice after comprehensive consideration from multiple dimensions. This decision-making mechanism based on maximizing the comprehensive evaluation value avoids the local optimum or suboptimal problem that may be caused by selecting a path based on a single indicator, thereby significantly improving the global optimization level of path planning. Specifically, it enables UAVs to more accurately and efficiently identify the optimal flight path that balances safety, flight efficiency, energy consumption, and regulatory compliance when facing complex and changing environments, effectively reducing flight risks and improving the reliability and economy of mission completion.

[0057] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for dynamic path planning and obstacle avoidance of unmanned aerial vehicles (UAVs) based on real-time environmental perception, characterized in that, The method specifically includes: Obtain real-time environmental data of the drone's current flight path, and determine whether the drone's current flight path needs to be replanned based on the real-time environmental data of the drone's current flight path; If the current flight path of the drone needs to be replanned, several alternative paths will be generated based on the current location of the drone. Obtain path feature data of alternative paths, and generate collision risk values ​​for alternative paths based on the feature data of alternative paths; where path feature data refers to feature data in the flight path that may cause the drone to collide. Obtain estimated travel time, estimated energy consumption, and airspace rule data for alternative routes; A comprehensive evaluation model for alternative routes is established based on the estimated travel time, estimated energy consumption, airspace rule data, and collision risk values ​​of alternative routes, and a comprehensive evaluation value for alternative routes is generated. The optimal alternative path is generated based on the comprehensive evaluation of the alternative paths. Based on the optimal alternative path, the drone's path is planned and obstacle avoidance is performed.

2. The UAV dynamic path planning and obstacle avoidance method based on real-time environmental perception according to claim 1, characterized in that, The methods for determining whether the current flight path of the drone needs to be replanned specifically include: Based on real-time environmental data of the current flight path of the drone, a safety factor for the current flight path is generated; Based on the safety factor of the current flight path, determine whether the current flight path of the drone needs to be replanned.

3. The UAV dynamic path planning and obstacle avoidance method based on real-time environmental perception according to claim 2, characterized in that, The method for generating the safety factor of the current flight path specifically includes: Through the formula: ; Generate the safety factor of the current flight path ; In the formula, This represents the normalized value of the i-th real-time environmental data point along the current flight path of the drone. This represents the weighting coefficient of the i-th real-time environmental data.

4. The UAV dynamic path planning and obstacle avoidance method based on real-time environmental perception according to claim 1, characterized in that, The specific methods for generating the collision risk value of the alternative paths include: Through the formula: ; Generate collision risk values ​​for alternative paths ; In the formula, F represents the frequency of historical environmental mutations in the candidate paths. Minimum frequency of historical environmental mutations in alternative paths This represents the maximum frequency of historical environmental mutations among the alternative paths. This represents the normalized value of the feature data of the j-th path among the candidate paths. This represents the weight coefficient of the j-th path feature among the candidate paths. , All are weighting ratio coefficients, and .

5. The UAV dynamic path planning and obstacle avoidance method based on real-time environmental perception according to claim 1, characterized in that, The specific methods for generating the comprehensive evaluation value of the alternative paths include: Based on the estimated travel time of the alternative routes, a timeliness loss factor is generated; Based on the estimated energy consumption of alternative routes, a range impact factor is generated; Generate a legality assessment factor based on the airspace rule data of the alternative paths; A comprehensive evaluation model for alternative routes is established based on timeliness loss factors, range impact factors, legality assessment factors, and alternative route collision risk values, generating a comprehensive evaluation value for alternative routes.

6. The UAV dynamic path planning and obstacle avoidance method based on real-time environmental perception according to claim 5, characterized in that, The specific methods for generating the timeliness loss factor include: Through the formula: ; Generate timely loss factor ; In the formula, This indicates the estimated travel time for the alternative routes. This indicates the estimated travel time of the drone's current flight path.

7. The UAV dynamic path planning and obstacle avoidance method based on real-time environmental perception according to claim 5, characterized in that, The specific methods for generating the range impact factor include: Through the formula: ; Generate battery life influencing factors ; In the formula, This indicates the estimated energy consumption of the alternative routes. This indicates the estimated energy consumption of the drone's current flight path.

8. The UAV dynamic path planning and obstacle avoidance method based on real-time environmental perception according to claim 5, characterized in that, The specific methods for generating the legality assessment factors include: Through the formula: ; Generate a legality assessment factor L; In the formula, This represents the k-th spatial rule data of the candidate paths. For conditional flag functions, This represents the weight coefficient of the k-th spatial rule.

9. The UAV dynamic path planning and obstacle avoidance method based on real-time environmental perception according to claim 5, characterized in that, The expression for the comprehensive evaluation model of the alternative paths is: ; In the expression, Score represents the overall evaluation value of the alternative paths. This represents the collision risk value of the alternative paths. This represents the timeliness loss factor. This represents the range-affecting factor, and L represents the legality assessment factor. , , All are weighting ratio coefficients, and .

10. The UAV dynamic path planning and obstacle avoidance method based on real-time environmental perception according to claim 1, characterized in that, The specific methods for generating the optimal alternative path include: Through the formula: ; Generate the optimal alternative path ; In the formula, This represents the g-th alternative path. This represents the comprehensive evaluation value of the candidate path for the g-th candidate path. This represents the set of all alternative paths.

Images