Sine Jerk Curve Robot Dynamic Grasping Trajectory Planning Method and System

Abstract

The present invention belongs to the field of industrial robot trajectory planning and specifically discloses a sinusoidal jerk curve robot dynamic grasping trajectory planning method and system. The method comprises: dividing the robot jerk trajectory curve into 15 segments, including 7 acceleration segments in the front, 1 uniform speed segment in the middle, and 7 deceleration segments in the back, and describing the trajectory curve using a time parameter; thereby obtaining a trajectory planning model; determining decision variables for the sinusoidal jerk curve trajectory planning model; determining constraints for the decision variables; establishing a fitness function based on the decision variable constraints, and using a genetic algorithm to solve the optimal solution for the decision variables; substituting the optimal solution for the decision variables into the trajectory planning model to obtain the robot dynamic grasping trajectory. The present invention can complete the pursuit of a uniformly moving workpiece in the shortest time, and has the conditions to track the moving workpiece at a uniform speed after catching up with the moving workpiece. At the same time, it achieves a continuous and smooth trajectory at the jerk level, reducing vibration and impact.

Description

Technical Field

[0001] The present invention belongs to the field of industrial robot trajectory planning, and more specifically, relates to a method and system for planning a sinusoidal jerk curve robot dynamic grasping trajectory. Background Art

[0002] With the vigorous development of the manufacturing industry and the surge in labor costs, the manufacturing industry is gradually transforming and upgrading towards digitalization, unmanned operation and human-machine collaboration. Robots have the advantages of high automation level, sustainable operation and flexible deployment, and have become the main focus of the transformation and upgrading of the manufacturing industry.

[0003] Trajectory planning is one of the key technologies in robot motion control, determining the robot's operating mode and performance. When grasping static workpieces, the motion trajectory must be kept smooth and continuous while satisfying kinematic constraints. When grasping dynamic workpieces, the runtime should be optimized to ensure grasping efficiency while ensuring kinematic constraints and trajectory continuity. However, most current trajectory planning methods only achieve continuous and smooth trajectories at the acceleration level, which can produce certain vibrations and impacts. While the sinusoidal jerk curve trajectory planning method allows the robot to move within the maximum kinematic constraints for as long as possible, reducing the total execution time, the runtime for acceleration and deceleration phases is the same, making it only suitable for robots grasping static workpieces and not for grasping dynamic workpieces. Summary of the Invention

[0004] In response to the above-mentioned defects or improvement needs of the prior art, the present invention provides a method and system for dynamic grasping trajectory planning of a robot with a sinusoidal jerk curve, the purpose of which is to ensure that the robot's acceleration is continuous while the speed at the end point is not zero, reduce vibration and impact, and improve the robot's motion smoothness and motion accuracy.

[0005] To achieve the above objectives, according to a first aspect of the present invention, a method for dynamic grasping trajectory planning of a robot with a sinusoidal jerk curve is proposed, comprising the following steps:

[0006] S1. Divide the robot's acceleration trajectory into 15 segments, including the first 7 acceleration segments, the middle 1 constant speed segment, and the last 7 deceleration segments. The trajectory curve is described by time parameters T1 to T7. Specifically:

[0007]

[0008] Among them, τ i =tt i , which represents the current time t to time node t i relative time, i = 0, 1, ... 15; J(t) is the acceleration trajectory curve; is the maximum absolute value of the acceleration during the acceleration phase, is the maximum value of the absolute value of the acceleration in the deceleration stage; T1 = t1-t0 = t3-t2 = t5-t4 = t7-t6; T2 = t2-t1 = t6-t5; T3 = t4-t3; T4 = t8-t7; T5 = t9-t8 = t 11 -t 10 =t 13 -t 12 =t 15 -t 14 ; T6 = t 10 -t9=t 14 -t 13 ; T7 = t 12 -t 11 ;

[0009] Based on the jerk trajectory curve, a sinusoidal jerk curve trajectory planning model is obtained;

[0010] S2. Determine the decision variables of the sinusoidal acceleration curve trajectory planning model;

[0011] S3. Determine the constraints of decision variables;

[0012] S4. Establish a fitness function based on the decision variable constraints; based on the fitness function, use a genetic algorithm to solve the optimal solution of the decision variable according to the relative position of the robot and the workpiece to be grasped and the speed of the workpiece;

[0013] S5. Substitute the optimal solution of the decision variables into the sinusoidal acceleration curve trajectory planning model to obtain the robot's dynamic grasping trajectory.

[0014] As a further preferred embodiment, in step S1, based on the jerk trajectory curve, the jerk, acceleration, and velocity are integrated in sequence to obtain a sinusoidal jerk curve trajectory planning model, and the formula is as follows:

[0015]

[0016] Where: parameter τ∈(t i ,t i+1 ], a(t) is the acceleration trajectory curve, v(t) is the velocity trajectory curve, and d(t) is the displacement trajectory curve.

[0017] As a further preferred embodiment, in step S2, since the time parameters T1 to T7 are not suitable to be directly used as decision variables, α1, α2 is used as the decision variable, and its expression relationship with the time parameter is determined as follows:

[0018]

[0019] in, are the maximum absolute values of acceleration in the acceleration phase and the deceleration phase respectively; is the maximum speed, is the workpiece speed, that is, the final speed of the robot; α1 and α2 are the first smoothing parameter and the second smoothing parameter respectively; Δd is the distance between the robot end position and the workpiece position.

[0020] As further preferred, in step S3, the constraint conditions of the decision variables include nonlinear inequality constraints and nonlinear equality constraints;

[0021] The nonlinear inequality constraints are:

[0022]

[0023] The nonlinear equality constraints are:

[0024]

[0025] Among them, J max 、A max 、V max are theoretical jerk, acceleration, and maximum velocity, respectively; d 15 is the total displacement of the robot during the grasping process, V belt is the moving speed of the workpiece during the grasping process, T sum The total time of the robot grasping process.

[0026] As a further preferred embodiment, in step S4, the fitness function f fitness for:

[0027]

[0028] Among them, T i Indicates the time required for the robot to complete a certain trajectory curve, ω i Indicates T i The corresponding weight, f max The maximum value that can be taken for a given total execution time range.

[0029] As a further preferred embodiment, in step S4, obtaining the optimal solution of the decision variables using a genetic algorithm includes:

[0030] (1) Initialize the population so that each individual in the population corresponds to a solution to a set of decision variables;

[0031] (2) Calculate the fitness value of each individual in the current population according to the fitness function;

[0032] (3) Based on the fitness value, the individuals in the population are selected, crossed, and mutated to obtain the next generation population;

[0033] (4) Repeat steps (2) and (3) until the stopping requirements of the genetic algorithm are met and the optimal solution of the decision variables is obtained.

[0034] As a further preferred method, the selection, crossover and mutation operations on the individuals in the population are specifically as follows:

[0035] Select the operation mode: set the number of individuals in the population to N and the fitness of the individual to f x , then the probability of each individual being selected is

[0036] Crossover operation method: Let e1 and e2 be the two selected excellent individuals, and generate two new individuals e'1 and e'2 through the crossover operator. The formula is as follows:

[0037]

[0038] Among them, λ is the crossover probability, which ranges from (0,1];

[0039] Mutation operation method: Perform mutation operation on excellent individuals. The mutation formula is:

[0040]

[0041] Among them, e is an excellent individual, that is, a mutant gene; g is the current generation, GEN is the maximum generation, μ1 and μ2 are random numbers, and e max and e min Indicates the maximum and minimum values of the gene.

[0042] According to a second aspect of the present invention, a computer-readable storage medium is provided for storing computer instructions, which, when executed by a processor, implement the above-mentioned robot dynamic grasping trajectory planning method.

[0043] According to a third aspect of the present invention, a robot dynamic grasping trajectory planning system is provided, comprising a processor, wherein the processor is configured to execute the above-mentioned robot dynamic grasping trajectory planning method.

[0044] In general, the above technical solutions conceived by the present invention have the following technical advantages compared with the existing technology:

[0045] 1. This invention divides the jerk trajectory curve during the robot's grasping process into 15 segments and describes the trajectory using time parameters. Seven time parameters represent seven motion modes. These are independently set and do not interfere with each other. The corresponding time parameters enable a smooth transition between the acceleration and velocity curves, and a genetic algorithm is then used to solve the decision variables. Through the design of the jerk trajectory curve, this invention can complete the pursuit of a uniformly moving workpiece in the shortest possible time. After catching up with the moving workpiece, it is able to track the moving workpiece at a uniform speed. Simultaneously, it achieves continuous and smooth trajectory at the jerk level, reducing vibration and impact.

[0046] 2. Since the time parameter is not suitable to be used directly as a decision variable, the present invention cleverly designs the decision variables related to the time parameter and its constraints. On the premise of satisfying the kinematic constraints of the robot, it can not only set the target displacement and terminal velocity of the motion trajectory, but also has a strong time optimization capability. At the same time, it maximizes the use of joint performance and can obtain a smooth and continuous motion trajectory at the acceleration level, reducing vibration and impact, improving motion smoothness and motion accuracy, and having excellent motion performance.

[0047] 3. Since if the trajectory obtained by the trajectory planning algorithm in Cartesian space is continuous, the trajectory of the controlled robot in the joint space is also continuous, and vice versa, the present invention is applicable to the trajectory planning of robots in Cartesian space and joint space. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 This is a flow chart of a method for planning a dynamic grasping trajectory of a robot with a sinusoidal acceleration curve based on a genetic algorithm, provided by an embodiment of the present invention;

[0049] Figure 2 Schematic diagram of a sinusoidal jerk curve trajectory planning method model provided by an embodiment of the present invention, wherein (a) to (d) are the jerk trajectory curve, the acceleration trajectory curve, the velocity trajectory curve, and the displacement trajectory curve, respectively;

[0050] Figure 3 It is a physical model of a robot end-user pursuing a moving workpiece provided by an embodiment of the present invention;

[0051] Figure 4 This is a time optimization flow chart of a sinusoidal acceleration curve method based on a genetic algorithm provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0053] The embodiment of the present invention provides a sinusoidal acceleration curve robot dynamic grasping trajectory planning method, such as Figure 1 As shown, the following steps are included:

[0054] Step S1: Establish a sinusoidal jerk curve trajectory planning model.

[0055] The trajectory planning model of the sinusoidal acceleration curve is as follows Figure 2 As shown, it includes a jerk trajectory curve, an acceleration trajectory curve, a velocity trajectory curve and a displacement trajectory curve.

[0056] In order to achieve the predetermined position and velocity while making the trajectory transition as smooth as possible, the trajectory curve of the jerk is divided into 15 segments, including the first 7 acceleration segments, the middle 1 constant speed segment, and the last 7 deceleration segments.

[0057] The 15 trajectories can be divided into 4 non-uniform acceleration stages, 2 uniform acceleration stages, 1 uniform acceleration stage, 1 uniform speed stage, 4 non-uniform deceleration stages, 2 uniform deceleration stages and 1 uniform deceleration stage; the same stages have the same time. Figure 2 , from the beginning to the end of the 15 segments, segments 1, 3, 13, and 15 are non-uniform jerk stages, segments 2 and 14 are uniform jerk stages, segment 4 is uniform acceleration stage, segment 8 is uniform speed stage, segments 5, 7, 9, and 11 are non-uniform deceleration stages, segments 6 and 10 are uniform deceleration stages, and segment 12 is uniform deceleration stage. More specifically, the jerk trajectory curve J(t) is as follows:

[0058]

[0059] Where: τ i =tt i , indicating t to t i relative time; is the maximum absolute value of the acceleration during the acceleration phase, It is the maximum absolute value of the acceleration during the deceleration phase.

[0060] The trajectory is described by time parameters T1 to T7, where T1 = t1-t0 = t3-t2 = t5-t4 = t7-t6; T2 = t2-t1 = t6-t5; T3 = t4-t3; T4 = t8-t7; T5 = t9-t8 = t 11 -t 10 =t 13 -t 12 =t 15 -t 14 ; T6 = t 10 -t9=t 14 -t 13 ; T7 = t 12 -t 11 ; The correspondence of this time parameter makes the acceleration curve and the speed curve transition smoothly.

[0061] Then, the sinusoidal jerk curve trajectory planning model is obtained by integrating the jerk, acceleration, and velocity in sequence. The derivation formula is as follows:

[0062]

[0063] Where: τ∈(t i ,t i+1 ], a(t) is the acceleration trajectory curve, v(t) is the velocity trajectory curve, and d(t) is the displacement trajectory curve.

[0064] Specifically, seven time parameters T1 to T7 are used to describe the trajectory. According to the above formula, the trajectory curve equation group of the specific acceleration, velocity and displacement of each segment of the trajectory can be obtained:

[0065] S11 When time t∈(t0,t1], time interval ΔT=T1, initial acceleration a0, velocity v0 and displacement d0 are all set to 0, the trajectory equation is:

[0066]

[0067] S12 When time t∈(t1,t2], time interval ΔT=T2, the trajectory planning equation is:

[0068]

[0069] Where:

[0070] S13 When time t∈(t2,t3], time interval ΔT=T1, the trajectory planning equation is:

[0071]

[0072] Where:

[0073] S14 When time t∈(t3,t4], time interval ΔT=T3, the trajectory planning equation is:

[0074]

[0075] Where:

[0076] S15 When time t∈(t4,t5], time interval ΔT=T1, the trajectory planning equation is:

[0077]

[0078] Where:

[0079] S16 When time t∈(t5,t6], time interval ΔT=T2, the trajectory planning equation is:

[0080]

[0081] Where:

[0082] S17 When time t∈(t6,t7], time interval ΔT=T1, the trajectory planning equation is:

[0083]

[0084] Where:

[0085] S18 When time t∈(t7,t8], time interval ΔT=T4, the trajectory planning equation is:

[0086]

[0087] Where:

[0088] When time t∈(t8,t9], time interval ΔT=T5, the trajectory planning equation is:

[0089]

[0090] Where: a8 = 0;

[0091] S19 When time t∈(t9,t 10 ], the time interval ΔT=T6, and the trajectory planning equation is:

[0092]

[0093] Where:

[0094] S110When time t∈(t 10 ,t 11 ], the time interval ΔT=T5, and the trajectory planning equation is:

[0095]

[0096] Where:

[0097] S111When time t∈(t 11 ,t 12 ], the time interval ΔT=T7, and the trajectory planning equation is:

[0098]

[0099] Where:

[0100] S112 When time t∈(t 12 ,t 13 ], the time interval ΔT=T5, and the trajectory planning equation is:

[0101]

[0102] Where:

[0103] S113 When time t∈(t 13 ,t 14 ], the time interval ΔT=T6, and the trajectory planning equation is:

[0104]

[0105] Where:

[0106] S114 When time t∈(t 14 ,t 15 ], the time interval ΔT=T5, and the trajectory planning equation is:

[0107]

[0108] Where:

[0109] S115 motion trajectory endpoint acceleration, velocity and displacement equations:

[0110]

[0111] Step S2: Analyze the sinusoidal jerk curve trajectory planning model and determine the decision variables of the model.

[0112] S21 inputs the workpiece starting position d piece , robot starting position d tip and workpiece speed

[0113] S22 combines the formula derived in step S1 to obtain the following equations for kinematic parameters and time parameters:

[0114]

[0115] Assumptions As we know, there are 5 sets of equations and 8 unknowns in step S22, so three sets of equations are needed to solve the equations and solve the time parameter T. i (i=1,2…7).

[0116] S23 Since the terminal velocity obtained by the sinusoidal acceleration curve trajectory planning algorithm is not zero, and the displacement reaches the set value at the end of the movement, a physical model of movement in a certain direction is constructed by taking the pursuit of the moving workpiece as an example, such as Figure 3 As shown, at the initial moment t0, the robot is Δd away from the moving workpiece, and after T sum After a certain time, the two meet, so the formula is established:

[0117] d 15 +Δd=V belt T sum =V belt (4T1+2T2+T3+T4+4T5+2T6+T7)

[0118] Where: Δd = d tip -dpiece , represents the initial position d of the robot end tip and the workpiece initial position d piece distance.

[0119] S24 combined with steps S22 and S23 still cannot solve the time parameter. Two more sets of equations are needed to uniquely determine the time parameter. Therefore, combined with the smoothness requirement of the acceleration trajectory, two additional sets of equations are established:

[0120]

[0121] Combining the above three steps S22 to S24, the time parameter solution set is determined as follows:

[0122]

[0123] S25 Since the robot motion is affected by kinematic constraints, terminal velocity and set displacement value, the time parameters T1 to T7 are not suitable to be directly used as decision variables, and the time parameters can be obtained by α1, α2 is fully expressed, so it is determined that α1, α2 is used as a decision variable.

[0124] Step S3: Determine the constraint conditions of the decision variables by combining the kinematic constraints, target displacement, and target velocity.

[0125] S31 input kinematic constraint parameters: theoretical maximum acceleration J max , Theoretical maximum acceleration A max , Theoretical maximum speed V max ;

[0126] S32 further defines the nonlinear constraints of the decision variables according to the workpiece movement speed and the relative position of the workpiece and the robot starting point;

[0127] (1) If you can determine whether to accelerate or decelerate first to catch up with the object, then:

[0128] Case 1: When When Δd>0, the robot end is in the positive direction of the grasped object and the object's velocity is in the negative direction. Therefore, the robot end needs to decelerate and then accelerate to catch up with the object in the shortest time. The inequality equations are as follows:

[0129]

[0130] Case 2: When When Δd < 0, the end of the robot is in the opposite direction of the object being grasped and the object's velocity is in the opposite direction. Therefore, it is completely opposite to the first situation. It is necessary to accelerate first and then decelerate to catch up with the object in the shortest time. The inequality equations are as follows:

[0131]

[0132] The maximum kinematic constraints in the acceleration phase and the deceleration phase are different. This is to ensure that the terminal velocity reaches the set value and the predetermined target position can be reached.

[0133] (2) If it is not certain whether to accelerate or decelerate first to catch up with the object, it is necessary to rely on the genetic algorithm to make a judgment. The inequality constraint equations are as follows:

[0134]

[0135] S33 determines nonlinear equality constraints on decision variables:

[0136]

[0137] In step S4, a fitness function is established in combination with the decision variable constraint conditions obtained in step S3, and a genetic algorithm is used to solve the value of the decision variable.

[0138] S41 establishes the time fitness function equation:

[0139]

[0140] Where, T i Indicates the time required for the robot to complete a certain trajectory curve, ω i Represents the weight, that is, its value is equal to T i The proportion of the corresponding motion stage in the entire motion trajectory, f max Indicates the maximum value that can be taken for a given total execution time range.

[0141] S42 sets the selection operation mode of the genetic algorithm:

[0142] The selection operation of the genetic algorithm adopts the roulette method, that is, the larger the fitness value of the individual, the greater the probability of being selected. If the number of individuals in the population is set to N and the fitness of the individual is f x , then the probability of each individual being selected is:

[0143]

[0144] S43 sets the crossover operation mode of the genetic algorithm:

[0145] The crossover operator adopts arithmetic form. Let e1 and e2 be two selected excellent individuals. The crossover operator is used to generate two new individuals e'1 and e'2. The formula is as follows:

[0146]

[0147] Where λ is the crossover probability, which ranges from (0,1].

[0148] S44 sets the mutation operation mode of the genetic algorithm:

[0149] In order to prevent the genetic algorithm from falling into local convergence, it is necessary to perform mutation operations on outstanding individuals. The mutation formula is:

[0150]

[0151] Where e is the mutant gene, g is the current generation, GEN is the maximum generation, μ1 and μ2 are random numbers, and e max and e min Indicates the maximum and minimum values of the gene.

[0152] S45 uses genetic algorithm to obtain the solution of decision variables. The calculation process is as follows: Figure 4 As shown, including:

[0153] S451 sets the parameters of the genetic algorithm, such as the initial population size, the initial population value range, the maximum genetic generation number, and the crossover probability;

[0154] S452 calculates the time fitness value according to the nonlinear equality constraint and the nonlinear inequality constraint of the decision variable;

[0155] S453 determines whether the stopping requirement of the genetic algorithm is met. If so, the process jumps to step S455. If not, the process jumps to step S454. S454 performs selection, crossover, and mutation operations on the individuals in the population to obtain a new population, and repeats steps S452 and S453.

[0156] S455 outputs the optimal solution of the decision variables.

[0157] Specifically, the stopping requirement means that in multiple rounds of cycles, there exists an individual whose fitness value is always the largest or meets a certain specified requirement, then the individual with the largest fitness is the optimal solution.

[0158] Step S5: bringing the values of the decision variables into the sinusoidal jerk curve trajectory planning model to uniquely determine the sinusoidal jerk curve trajectory equation.

[0159] The method of the present invention uses a sinusoidal function curve to model the jerk motion trajectory, and obtains the motion trajectory of acceleration, velocity and displacement through the integration method, thereby obtaining a mathematical model of the sinusoidal jerk curve trajectory planning method; then, a genetic algorithm is used to determine the decision variables of the sinusoidal jerk curve trajectory planning algorithm in combination with the kinematic constraints. Under the premise of satisfying the kinematic constraints of the robot, the method of the present invention can not only set the target displacement and terminal velocity of the motion trajectory, but also has a strong time optimization capability. At the same time, a smooth and continuous motion trajectory can be obtained at the jerk level, which has certain advantages in reducing the vibration and impact of the robot movement; at the same time, the present invention is suitable for the trajectory planning of the robot in Cartesian space and joint space.

[0160] It will be easily understood by those skilled in the art that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A sinusoidal acceleration curve robot dynamic grasping trajectory planning method, characterized in that: The steps include: S1. Divide the robot's acceleration trajectory into 15 segments, including the first 7 acceleration segments, the middle 1 constant speed segment, and the last 7 deceleration segments. The trajectory curve is described by time parameters T1 to T7. Specifically: Among them, τ i =tt i , which represents the current time t to time node t i relative time, i = 0, 1, ... 15; J(t) is the acceleration trajectory curve; is the maximum absolute value of the acceleration during the acceleration phase, is the maximum value of the absolute value of the acceleration in the deceleration stage; T1 = t1-t0 = t3-t2 = t5-t4 = t7-t6; T2 = t2-t1 = t6-t5; T3 = t4-t3; T4 = t8-t7; T5 = t9-t8 = t 11 -t 10 =t 13 -t 12 =t 15 -t 14 ; T6 = t 10 -t9=t 14 -t 13 ; T7 = t 12 -t 11 ; Based on the jerk trajectory curve, a sinusoidal jerk curve trajectory planning model is obtained; S2. Determine the decision variables of the sinusoidal acceleration curve trajectory planning model; In step S2, since the time parameters T1 to T7 are not suitable to be directly used as decision variables, α1 α2 is used as the decision variable, and its expression relationship with the time parameter is determined as follows: in, are the maximum absolute values of acceleration in the acceleration phase and the deceleration phase respectively; is the maximum speed, is the workpiece speed, that is, the final speed of the robot; α1 and α2 are the first smoothing parameter and the second smoothing parameter respectively; Δd is the distance between the robot end position and the workpiece position; S3. Determine the constraints of decision variables; S4. Establish a fitness function based on the decision variable constraints; based on the fitness function, use a genetic algorithm to solve the optimal solution of the decision variable according to the relative position of the robot and the workpiece to be grasped and the speed of the workpiece; S5. Substitute the optimal solution of the decision variables into the sinusoidal acceleration curve trajectory planning model to obtain the robot's dynamic grasping trajectory.

2. The sinusoidal jerk curve robot dynamic grasping trajectory planning method according to claim 1, characterized in that: In step S1, based on the jerk trajectory curve, the jerk, acceleration, and velocity are integrated in sequence to obtain a sinusoidal jerk curve trajectory planning model, which is as follows: Where: parameter τ∈(t i , t i+1 ], a(t) is the acceleration trajectory curve, v(t) is the velocity trajectory curve, and d(t) is the displacement trajectory curve.

3. The sinusoidal jerk curve robot dynamic grasping trajectory planning method according to claim 1, characterized in that: In step S3, the constraints of the decision variables include nonlinear inequality constraints and nonlinear equality constraints; The nonlinear inequality constraints are: The nonlinear equality constraints are: Among them, J max 、A max 、V max are theoretical jerk, acceleration, and maximum velocity, respectively; d 15 is the total displacement of the robot during the grasping process, V belt is the moving speed of the workpiece during the grasping process, T sum The total time of the robot grasping process.

4. The sinusoidal jerk curve robot dynamic grasping trajectory planning method according to claim 1, characterized in that: In step S4, the fitness function f fitness for: Among them, T i Indicates the time required for the robot to complete a certain trajectory curve, ω i Indicates T i The corresponding weight, f max The maximum value that can be taken for a given total execution time range.

5. The sinusoidal jerk curve robot dynamic grasping trajectory planning method according to claim 4, characterized in that: In step S4, the optimal solution of the decision variables is obtained using a genetic algorithm, including: (1) Initialize the population so that each individual in the population corresponds to a solution to a set of decision variables; (2) Calculate the fitness value of each individual in the current population according to the fitness function; (3) Based on the fitness value, the individuals in the population are selected, crossed, and mutated to obtain the next generation population; (4) Repeat steps (2) and (3) until the stopping requirements of the genetic algorithm are met and the optimal solution of the decision variables is obtained.

6. The sinusoidal jerk curve robot dynamic grasping trajectory planning method according to claim 5, characterized in that: The selection, crossover, and mutation operations on population individuals are as follows: Select the operation mode: set the number of individuals in the population to N and the fitness of the individual to f x , then the probability of each individual being selected is Crossover operation method: Let e1 and e2 be the two selected excellent individuals, and generate two new individuals e'1 and e'2 through the crossover operator. The formula is as follows: Among them, λ is the crossover probability, which ranges from (0,1]; Mutation operation method: Perform mutation operation on excellent individuals. The mutation formula is: Among them, e is an excellent individual, that is, a mutant gene; g is the current generation, GEN is the maximum generation, μ1 and μ2 are random numbers, and e max and e min Indicates the maximum and minimum values of the gene.

7. A computer-readable storage medium for storing computer instructions, characterized in that: When the computer instructions are executed by the processor, the robot dynamic grasping trajectory planning method according to any one of claims 1 to 6 is implemented.

8. A robot dynamic grasping trajectory planning system, characterized in that: The method comprises a processor configured to execute the robot dynamic grasping trajectory planning method according to any one of claims 1 to 6.

Images