A multi-objective optimization-based lightweight autonomous task planning and scheduling method for double-star formation satellite

Abstract

A kind of double star formation satellite lightweight autonomous task planning and scheduling method based on multi-objective optimization, the method based on on-board constraint, realize the comprehensive performance optimal double star formation satellite autonomous task planning and scheduling, ensure the high-precision inter-satellite pointing and inter-satellite distance keeping during long-term operation of double star.This application takes the distance between double stars and the execution ability of satellite thruster as constraints, establishes a mathematical model of task planning, generates formation reference star scheduling and orbit / attitude control strategy based on multi-optimization objectives, realizes the full autonomy, lightweight, comprehensive performance optimal rapid autonomous task scheduling on board.

Description

TECHNICAL FIELD

[0001] The application relates to a lightweight autonomous task planning and scheduling method for a double-satellite formation based on multi-objective optimization, and belongs to the technical field of spacecraft control. BACKGROUND

[0002] For a low-low tracking gravity measurement satellite, double-satellite formation needs to be operated for a long time, and inter-satellite pointing attitude control and inter-satellite distance maintaining orbit control need to be realized during the operation, the process involves a complex task sequence, and the attitude and orbit control accuracy is high.

[0003] In order to improve the autonomous survival ability of the satellite, based on the on-satellite constraints, a double-satellite formation autonomous task planning and scheduling method with optimal comprehensive performance is designed. The inter-satellite pointing target angle and the inter-satellite distance are calculated autonomously by using the double-satellite orbit parameters provided by the ground or the double-satellite orbit information provided by the inter-satellite communication. The task planning mathematical model is established by taking the double-satellite distance and the execution ability of the satellite thruster as constraints. The formation reference satellite scheduling and orbit control strategy are generated based on multi-optimization objectives, and the high-precision inter-satellite pointing control is autonomously switched in after the orbit control ends. The optimal attitude control task sequence and configuration parameters are selected according to the terminal input of the orbit control. SUMMARY

[0004] The application provides a lightweight autonomous task planning and scheduling method for a double-satellite formation based on multi-objective optimization. The double-satellite pointing performance and the inter-satellite distance index in the formation process are comprehensively considered. The formation reference satellite scheduling and orbit / attitude control strategy are generated based on multi-optimization objectives considering the on-satellite resource constraints. The lightweight, autonomous and optimal comprehensive performance rapid autonomous task scheduling on the satellite is realized.

[0005] The technical scheme of the application is as follows: a lightweight autonomous task planning and scheduling method for a double-satellite formation based on multi-objective optimization, comprising the following steps:

[0006] The inter-satellite pointing target angle and the inter-satellite distance are calculated by using the double-satellite orbit parameters provided by the ground or the double-satellite orbit information provided by the inter-satellite communication, and the calculated values are taken as the input of the task planning.

[0007] The double-satellite pointing control is performed in the non-orbit control stage or the position control is performed in the orbit control stage, and the task planning model is established.

[0008] On the basis of the task planning model, the attitude / orbit task strategy is generated by dynamic optimization, that is, the optimal scheduling scheme of the task sequence is obtained.

[0009] Preferably, the calculation formula of the inter-satellite pointing target angle θ r and ψ r is as follows:

[0010]

[0011] r pvo =A OI r pv

[0012] Normalization processing r pvo =r pvo / ||r pvo ||

[0013]

[0014] ψ r =arctan2(r pvo [1],r pvo [0])

[0015] Where x, y, and z represent the three-dimensional positions of the local star in the inertial frame of reference, respectively. T y T z T Let A and B represent the three-dimensional positions of the target star in the inertial frame, respectively. OI This is the direction cosine matrix of the local orbital system relative to the inertial frame;

[0016] || represents the norm of the matrix.

[0017] Preferably, the formula for calculating interstellar distance is:

[0018] D r =||r pv ||.

[0019] Preferably, dual-satellite pointing control is performed during the non-orbit control phase, or position control is performed during the orbit control phase. The target characteristics are analyzed and optimized, and a mission planning model is established.

[0020] minf = g(E1,E2,...,E n (1)

[0021]

[0022] Equation (1): is the objective function for planning and optimization, E i The total benefit of the formation and pointing of satellite i is represented by i∈{1,...,n}, where n is the total number of satellites in the formation. The multiplicative weighted method is used to integrate the optimization objectives.

[0023] Equation (2): Decision variables Let represent the decision variable for the task sequence j corresponding to the optimization objective i. If the task is executed, then... It is 1 if it is true, otherwise it is 0;

[0024] Equation (3): Decision variables A corresponding constraint set C, the constraint set C includes thruster thrust constraints, attitude constraints of the satellite;

[0025] Equation (4): The execution start time of the task sequence k corresponding to the optimization target i, w i The time zero point of the task sequence corresponding to the optimization target.

[0026] Preferably, the evaluation function of the weighted product or the weighted sum or the combination of both is used to simplify the solution of the task planning model.

[0027] Preferably, the evaluation function of the combination of the weighted product and the weighted sum is in the form of the following formalized model:

[0028]

[0029] In the formula, η i1 The weight coefficient of the task priority, the higher the priority, the greater the value; Represent the total length of the thruster start-up, Represent the total length of the inter-satellite distance maintenance; The number of task switching times; Represent the control accuracy; Represent the robustness index, that is, the tolerance envelope of the system to uncertainty; The corresponding weight coefficient, the value is between 0 and 1, and the sum of the total weight coefficients is 1.

[0030] A multi-objective optimization-based light-weight autonomous task planning and scheduling system for a double-satellite formation satellite, comprising:

[0031] An input determination module calculates the inter-satellite pointing target angle and predicts the inter-satellite distance using the double-satellite orbit parameters provided by the ground or the double-satellite orbit information provided by the inter-satellite communication, and uses the calculation value as the input of the task planning;

[0032] A model construction module establishes a task planning model by performing double-satellite pointing control in the non-orbit control stage or performing position control in the orbit control stage;

[0033] A dynamic optimization module uses the calculation results of the input determination module as the input of the task planning, and generates an attitude / orbit task strategy according to the established task planning model, that is, obtains the optimal scheduling scheme of the task sequence.

[0034] Preferably, the calculation formula of the inter-satellite pointing target angle θ r , ψ r is as follows:

[0035]

[0036] r pvo =AOI r pv

[0037] Normalization processing r pvo =r pvo / ||r pvo ||

[0038]

[0039] ψ r =arctan2(r pvo [1],r pvo [0])

[0040] Where x, y, and z represent the three-dimensional positions of the local star in the inertial frame of reference, respectively. T y T z T Let A and B represent the three-dimensional positions of the target star in the inertial frame, respectively. OI This is the direction cosine matrix of the local orbital system relative to the inertial frame;

[0041] || represents the norm of the matrix;

[0042] The formula for calculating interstellar distance is:

[0043] D r =||r pv ||.

[0044] Preferably, dual-satellite pointing control is performed during the non-orbit control phase, or position control is performed during the orbit control phase. The target characteristics are analyzed and optimized, and a mission planning model is established.

[0045] minf = g(E1,E2,...,E n (1)

[0046]

[0047] Equation (1): is the objective function for planning and optimization, E i The total benefit of the formation and pointing of satellite i is represented by i∈{1,...,n}, where n is the total number of satellites in the formation. The multiplicative weighted method is used to integrate the optimization objectives.

[0048] Equation (2): Decision variables Let represent the decision variable for the task sequence j corresponding to the optimization objective i. If the task is executed, then... It is 1 if it is true, otherwise it is 0.

[0049] Equation (3): Decision variables The corresponding constraint set C includes thrust constraints for the thruster and attitude constraints for the celestial body.

[0050] Equation (4): For the optimization target i corresponding to the starting time of the task sequence k, w i For the optimization target corresponding to the task sequence time zero point.

[0051] A computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the method for lightweight autonomous task planning and scheduling of a dual-satellite formation satellite based on multi-objective optimization.

[0052] Compared with the prior art, the present application has the following beneficial effects:

[0053] (1) The method for lightweight autonomous task planning and scheduling of a dual-satellite formation satellite based on multi-objective optimization proposed in the present application is a novel full-autonomous dynamic optimization solution method for low-low tracking formation satellite attitude and orbit cooperative control, which well solves the full-autonomous task planning problem of high-precision gravity measurement satellites that need to perform intersatellite pointing in a high-precision formation configuration. Similar methods of the present application have not been reported in domestic and foreign literature and public patents.

[0054] (2) Compared with the prior art, the technical method adopted in the present application fully exploits the task planning characteristics of low-low gravity tracking formation satellites, and through dynamic programming and optimal strategy design, it ingeniously solves the rapid and efficient task scheduling problem of dual-satellite formation, the algorithm is simple and feasible, and can meet the high-precision index requirements. The entire algorithm design is simple, and the parameter debugging workload is small.

[0055] (3) The present application proposes a new solution for the task planning problem of the entire process of low-low tracking gravity satellite formation, which does not need to increase additional data input, and the calculation is simple. This algorithm can be adapted to a large class of military and civil satellite systems with similar formation control requirements, and has strong engineering practicability. BRIEF DESCRIPTION OF DRAWINGS

[0056] Figure 1 The algorithm flowchart.

[0057] Figure 2 The intersatellite distance variation curve.

[0058] Figure 3 The pointing accuracy curve of the dual-satellite during non-orbit control. DETAILED DESCRIPTION

[0059] To make the purpose, technical scheme and advantages of the present application clearer, the following will combine the attached drawings to further describe the present application. Figure 1 The embodiments of the present application are further described in detail.

[0060] The application relates to a double-satellite formation satellite lightweight autonomous task planning and scheduling method based on multi-target optimization, which comprises the following steps of obtaining task planning input, a task planning model and a final optimization scheduling scheme respectively.

[0061] Step one, inter-satellite pointing target angle calculation and inter-satellite distance prediction calculation;

[0062] The double-satellite orbit extrapolation calculation, inter-satellite pointing target angle calculation and inter-satellite distance prediction calculation are autonomously performed by using double-satellite orbit parameters provided by the ground or double-satellite orbit information provided by inter-satellite communication.

[0063] The calculation formula of the inter-satellite pointing target angle theta r , and the inter-satellite distance D r .

[0064]

[0065] r pvo = A OI r pv

[0066] The normalized processing r pvo = r pvo / ||r pvo ||

[0067]

[0068] The calculation formula of the inter-satellite pointing target angle theta r = arctan2(r pvo [1], r pvo [0])

[0069] Wherein, x, y and z are three-dimensional positions of the satellite in the inertial system, x T , y T , z T are three-dimensional positions of the target satellite in the inertial system, and A OI is a direction cosine matrix of the satellite orbit system relative to the inertial system.

[0070] The calculation formula of the inter-satellite distance D r =||r pv ||

[0071]

[0072] Step two, task planning model establishment;

[0073] In the non-orbit control stage, double-satellite pointing control or position control is performed in the orbit control stage, the characteristics of optimization targets (the highest precision or the lowest fuel consumption or the highest sensor relaxation) are deeply analyzed, and a task planning model is established.

[0074] minf = g (E1, E2,..., En)​n ) (1)

[0075]

[0076] Equation (1) is a planning optimization objective function, E i represents the total revenue of the formation and pointing of satellite i, i∈{1,...,n}, n is the total number of formation satellites, and each optimization objective is integrated by using a multiplication weighting method;

[0077] Equation (2) is a decision variable represents the decision variable of the task sequence j corresponding to the optimization objective i, if the task is executed, then 1, otherwise 0;

[0078] Equation (3) is a decision variable corresponding constraint set C, such as thruster thrust constraint, attitude constraint of the star body, etc.

[0079] Equation (4): is the execution start time of the task sequence k corresponding to the optimization objective i, w i is the task sequence time zero point corresponding to the optimization objective.

[0080] To simplify the multi-objective optimization problem of formation and pointing control, a quantitative comprehensive evaluation function needs to be used to judge the pros and cons of the candidate tracking scheme when task allocation is performed, and an evaluation function of weighted product or weighted sum can be used. Here, an evaluation function combining the two is used, and the formalized description model is as follows:

[0081]

[0082] In the equation, E i represents the total revenue of the formation and pointing of satellite i. η i1 is the weight coefficient of the task priority, the higher the priority, the greater the value; represents the total length of time that the thruster is turned on, represents the total length of time that the inter-satellite distance is maintained; is the number of task switching times; represents the control accuracy; represents the robustness index, that is, the tolerance envelope of the system to uncertainty; is the corresponding weight coefficient, which is taken as a value between 0 and 1, and the sum of the total weight coefficients is 1. The specific value of each weight is configured in advance according to the characteristics of the task.

[0083] Step three, dynamic optimization is performed to generate and schedule the attitude / orbit task strategy.

[0084] On the basis of the formation task planning model, the optimal scheduling scheme is generated through hierarchical optimization. When the formation constraints (the phase difference of the current star and the target star and the interstellar distance meet the requirements) are satisfied, the satellite scheduling is the load task phase, and the attitude control is implemented according to the interstellar pointing attitude control mode; when the formation constraints are not satisfied, if the current star is the formation reference star, no orbit control is performed, and the attitude control is still implemented according to the interstellar pointing attitude control mode, and if the current star is not the formation reference star, the scheduling is the orbit control task phase, and the orbit control attitude maneuver mode is first implemented to establish the orbit control attitude, and then the interstellar distance maintenance orbit control mode is automatically switched to perform the orbit control operation.

[0085] The present application is to realize the on-orbit real-time fast calculation, and according to the specific optimization target (time optimization, fuel optimization, etc.), a hierarchical optimization strategy is adopted to divide the optimization process into rough evaluation optimization and fine evaluation optimization.

[0086] In the rough evaluation optimization, a rough evaluation model based on particle swarm genetic planning is established, and all parameter configuration schemes and task optional branches in the sample space are quickly evaluated to obtain a high-quality solution set, such as determining the formation reference star, sensor configuration and thrust configuration in the first stage. In the fine evaluation optimization, an iterative optimization method is adopted, the current parameters are modified through feedback information, the accuracy of the rough evaluation model is gradually improved through continuous optimization, and the size of the high-quality solution set is reduced, and on this basis, the time sequence of the load task phase and the orbit control task phase and the specific control configuration and control parameters are determined.

[0087] The present application also provides a double-satellite formation satellite lightweight autonomous task planning and scheduling system based on multi-objective optimization, characterized by comprising:

[0088] An input determination module is used to calculate the interstellar pointing target angle and the predicted interstellar distance by using the double-satellite orbit parameters provided by the ground or the double-satellite orbit information provided by the interstellar communication, and the calculation values are used as the input of the task planning;

[0089] A model construction module is used to establish a task planning model by performing double-satellite pointing control in the non-orbit control phase or performing position control in the orbit control phase;

[0090] A dynamic optimization module is used to generate an attitude / orbit task strategy by using the calculation results of the input determination module as the input of the task planning, and the optimal scheduling scheme of the task sequence is obtained according to the established task planning model.

[0091] The same parts in the specific functions of the modules and the methods can be referred to the related introduction in the methods.

[0092] The present application further provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to realize the steps of the above method.

[0093] The application will be further described in connection with the following examples.

[0094] Examples:

[0095] Taking a certain on-orbit satellite as an example, the whole-process autonomous planning method of the formation is simulated and verified. The formation configuration of the satellite double star in the simulation needs to make the inter-satellite distance as long as possible and stable at 170km-270km, and the satellite runs in the double star pointing mode during the non-orbit control period, and the pointing accuracy needs to meet ≤3mrad (3σ; pitch, yaw), ≤30mrad (3σ; roll). Taking the above index as the optimization target, the on-board autonomous task planning is carried out, and the satellite autonomously carries out twice double-pulse orbit control, the orbit control interval is 129 days, the inter-satellite distance control curve is as Figure 2 , and the double star pointing control accuracy curve is as Figure 3 .

[0096] Although the application has been disclosed with the above preferred embodiments, it is not intended to limit the application, and any person skilled in the art can make possible changes and modifications to the technical solutions of the application by using the disclosed methods and technical contents without departing from the spirit and scope of the application. Therefore, any simple modification, equivalent change and modification made to the above examples according to the technical essence of the application, which does not depart from the technical solutions of the application, belongs to the protection scope of the technical solutions of the application.

[0097] The contents not described in detail in the specification of the application are the known technology of those skilled in the art.

Claims

1. A method for lightweight autonomous task planning and scheduling of a dual-satellite formation based on multi-objective optimization, characterized in that It comprises: Using the double star orbit parameters provided by the ground or the double star orbit information provided by the interstellar communication, the interstellar pointing target angle and the predicted interstellar distance are calculated, and the calculated values are taken as the input of the task planning; In the non-orbit control stage, the double star pointing control or the position control in the orbit control stage is carried out, and the task planning model is established; On the basis of the task planning model, the dynamic optimization is adopted to generate the attitude / orbit task strategy, that is, the optimal scheduling scheme of the task sequence is obtained.

2. The lightweight autonomous task planning and scheduling method for dual-satellite formation based on multi-objective optimization according to claim 1, characterized in that, Inter-satellite pointing target angle θ r , ψ r The calculation formula is: r pvo ＝A OI r pv Utilizing r pvo Matrix norm normalization processing r pvo ψ r = arctan2(r pvo [1], r pvo [0]) Where x, y, and z represent the three-dimensional positions of the local star in the inertial frame of reference, respectively. T y T z T Let A and B represent the three-dimensional positions of the target star in the inertial frame, respectively. OI This is the direction cosine matrix of the local orbital system relative to the inertial frame.

3. The lightweight autonomous task planning and scheduling method for dual-satellite formation based on multi-objective optimization according to claim 2, characterized in that, The calculation formula of the interstellar distance is D r ＝||r pv || Wherein, || || is the norm of the matrix.

4. The lightweight autonomous task planning and scheduling method for dual-satellite formation based on multi-objective optimization according to claim 1, characterized in that, In the non-orbit control stage, the double star pointing control or the position control in the orbit control stage is carried out, the characteristics of the optimization target are analyzed, and the task planning model is established; minf = g (E1, E2,..., En) (1) n ) (1) Formula (1): is a planning optimization objective function, E i Total revenue of formation and pointing of satellite i, i∈{1,...,n}, n is the total number of formation satellites, and each optimization objective is integrated by using multiplication weighting method; Formula (2): decision variable denotes the decision variable of the task sequence j corresponding to the optimization target i, if the task is executed, then is 1, otherwise 0; Equation (3): decision variables Set of locations corresponding to a set of constraints C, the set of constraints C including thruster thrust constraints, attitude constraints of the space object; Formula (4): w is the execution start time of the task sequence k corresponding to the optimization target i i is the time zero point of the task sequence corresponding to the optimization target.

5. The lightweight autonomous task planning and scheduling method for dual-satellite formation based on multi-objective optimization according to claim 4, characterized in that, The evaluation function of the weighted product or the weighted sum or the combination of the two is used to simplify the solution of the task planning model.

6. The lightweight autonomous task planning and scheduling method for dual-satellite formation based on multi-objective optimization according to claim 5, characterized in that, The formalized description model of the evaluation function of the combination of the weighted product and the weighted sum is as follows: In the formula, η i1 is the weight coefficient of the task priority, the higher the priority, the greater the value; represents the total length of the thruster start-up, represents the total length of the inter-satellite distance maintenance; is the number of task switching times; represents the control accuracy; represents the robustness index, that is, the tolerance envelope of the system to uncertainty; η2~η6 are the corresponding weight coefficients, which are taken as the value between 0 and 1, and the sum of the total weight coefficients is 1.

7. A multi-objective optimization-based dual-satellite formation satellite lightweight autonomous task planning and scheduling system, characterized in that It comprises: The input determination module uses the double star orbit parameters provided by the ground or the double star orbit information provided by the interstellar communication to calculate the interstellar pointing target angle and the predicted interstellar distance, and takes the calculated values as the input of the task planning; The model construction module establishes the task planning model by carrying out the double star pointing control in the non-orbit control stage or the position control in the orbit control stage; The dynamic optimization module takes the calculation results of the input determination module as the input of the task planning, and generates the attitude / orbit task strategy according to the established task planning model, that is, the optimal scheduling scheme of the task sequence is obtained.

8. The lightweight autonomous task planning and scheduling system for dualsatellite formation based on multi-objective optimization according to claim 7, characterized in that, Inter-satellite pointing target angle θ r , ψ r The calculation formula is: r pvo ＝A OI r pv Utilizing r pvo Matrix norm normalization processing r pvo ψ r = arctan2(r pvo [1], r pvo [0]) Wherein, x, y, z are three-dimensional positions of the target star in the inertial system, respectively, and A is the direction cosine matrix of the target star orbit system relative to the inertial system. T T T OI Wherein, x, y, z are three-dimensional positions of the target star in the inertial system, respectively, and A is the direction cosine matrix of the target star orbit system relative to the inertial system.​​​ || || is the norm of the matrix; The calculation formula of the interstellar distance is D r ＝||r pv ||。 9. The lightweight autonomous task planning and scheduling system for dual satellite formation based on multi-objective optimization according to claim 7, characterized in that, In the non-orbit control stage, the double star pointing control or the position control in the orbit control stage is carried out, the characteristics of the optimization target are analyzed, and the task planning model is established; min f = g(E1, E2,..., E n ) (1) Formula (1): is a planning optimization objective function, E i Total revenue of formation and pointing of satellite i, i∈{1,...,n}, n is the total number of formation satellites, and each optimization objective is integrated by using multiplication weighting method; Formula (2): decision variable denotes the decision variable of the task sequence j corresponding to the optimization target i, if the task is executed, then is 1, otherwise 0; Equation (3): decision variables the set of which the user is located a corresponding set of constraints C, the set of constraints C including thruster thrust constraints, attitude constraints of the satellite; Formula (4): w is the execution start time of the task sequence k corresponding to the optimization target i i is the task sequence time zero point corresponding to the optimization target.

10. A computer-readable storage medium storing a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the method in any one of claims 1-6.

Images