Data-driven method for optimizing robot energy consumption
By obtaining the robot joint trajectory and performing variable time scale scaling and heuristic search, combined with dynamic programming algorithms, the energy consumption optimization model is built, and the problem of robot energy consumption calculation depends on technical parameters is solved, and the accurate optimization of energy consumption and evaluation of efficiency is achieved. improve.
Patent Information
- Application Number
- CN202210842327.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-07-18
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2042-07-18
AI Technical Summary
In the prior art, the calculation of robot energy consumption depends on detailed technical parameters, which is difficult for users to obtain, and the energy consumption changes with working conditions, making it difficult to accurately determine the robot energy consumption, especially at different feed rates, which vary greatly.
By obtaining the robot's preset joint trajectory, performing variable time scale scaling and meta-heuristic algorithm search, combining dynamic programming algorithms, building an energy consumption optimization calculation model, iteratively calculate the energy consumption of joint trajectory, and determining the optimal scaling trajectory and value of energy consumption without requiring Get other relevant parameters.
The accurate optimization and determination of robot energy consumption is achieved, the efficiency of energy consumption optimization evaluation is improved, and the accuracy of energy consumption calculation is ensured.
Smart Images

Figure BDA0003751621610000021 
Figure BDA0003751621610000024 
Figure BDA0003751621610000026
Abstract
Description
Technical Field
[0001] The present invention relates to a robot energy consumption optimization method, and in particular to a robot energy consumption optimization method based on data driving. Background Art
[0002] In the prior art, the energy consumption calculation of a robot relies on the detailed technical parameters of its drive system, including permanent magnet synchronous servo motors, frequency converters, and rectifiers. However, robot users often cannot obtain the corresponding parameters, and the robot's technical parameters will change with the working conditions. At the same time, for the same task, the energy consumption of the robot will vary greatly when different feed rates are used, making it difficult to accurately determine the energy consumption of the robot.
[0003] Therefore, in order to solve the above technical problems, it is urgent to propose a new technical means. Summary of the invention
[0004] In view of this, the purpose of the present invention is to provide a data-driven robot energy consumption optimization method, which can optimize and determine the energy consumption of the robot through the joint trajectory of the robot and the energy consumption value of executing the joint trajectory, without obtaining other relevant parameters of the robot, thereby effectively improving the efficiency of energy consumption optimization evaluation while ensuring the accuracy of energy consumption calculation.
[0005] The present invention provides a data-driven robot energy consumption optimization method, comprising the following steps:
[0006] S1. Obtain the preset joint trajectory q of the target robot r (u); where u is the execution time of the preset joint trajectory;
[0007] S2. Set the preset joint trajectory q of the target robot r (u) Divide the time domain into k segments and scale each joint trajectory with variable time scale; q i,s (t) = q i,r (u), 1≤i≤k; where the variable time scale scaling function is:
[0008]
[0009] Where: q s (t) is the joint trajectory after variable time scale scaling, t is the execution time of the joint trajectory after variable time scale scaling, C i are variable time scale scaling factors, It is a constant in the i-th time period;
[0010] S3. Use the meta-heuristic algorithm to search for n scaling joint trajectories that satisfy the robot motion constraints and time constraints and satisfy the full rank of the matrix A column, where the matrix A is:
[0011]
[0012] in:
[0013] S4. Control the target robot to execute the n scaling trajectories in order, and collect the energy consumption value E of each scaling trajectory in each time period i,m , where: 1≤i≤k, 1≤m≤n;
[0014] S5. Construct a calculation model for the robot's energy consumption characteristic parameters:
[0015] b=(A T A) -1 A T E, where: b = [b i,1 b i,2 … b i,12 ] T is the energy consumption characteristic parameter of the robot, E=[E i,1 E i,2 … E i,m … E i,n ] T ;
[0016] S6. Construct energy consumption optimization calculation model:
[0017]
[0018] ; Using dynamic programming algorithm, combined with the following time scaling state space expression to scale the parameters Search and iteratively calculate the energy consumption of the joint trajectory based on the energy consumption optimization calculation model and the state transfer equation, and finally obtain the optimal energy consumption scaling trajectory and the optimal energy consumption value that meet the specified conditions:
[0019]
[0020] Where: △ = u i+1 -u i , △ is a constant; the specified conditions are the position constraint, velocity constraint, acceleration constraint of the robot in the specified time period and the execution time of the entire trajectory after scaling;
[0021] In addition, the state transition equation of the dynamic programming algorithm is:
[0022]
[0023] Where: L irepresents the minimum joint trajectory energy consumption in the first i-1 time period, E i Indicates that the robot is in state Transfer to Energy consumption of joint trajectories calculated by the energy consumption optimization calculation model.
[0024] Furthermore, the robot motion constraints include position constraints, velocity constraints and acceleration constraints.
[0025] Furthermore, the time constraint is the maximum allowed time for the robot to execute the entire joint trajectory after scaling in step S2.
[0026] Furthermore, meta-heuristic algorithms include but are not limited to taboo search algorithm, simulated annealing algorithm, genetic algorithm, ant colony optimization algorithm, particle swarm optimization algorithm, artificial fish swarm algorithm, artificial bee colony algorithm, and artificial neural network algorithm.
[0027] Beneficial effects of the present invention: Through the present invention, the energy consumption of the robot can be optimized and determined through the joint trajectory of the robot and the energy consumption value obtained by executing the joint trajectory, without obtaining other relevant parameters of the robot, effectively improving the efficiency of energy consumption optimization evaluation while ensuring the accuracy of energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS
[0028] The present invention will be further described below in conjunction with the accompanying drawings and embodiments:
[0029] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION
[0030] The present invention is further described below:
[0031] The present invention provides a data-driven robot energy consumption optimization method, comprising the following steps:
[0032] S1. Obtain the preset joint trajectory q of the target robot r (u); where u is the execution time of the preset joint trajectory;
[0033] S2. Set the preset joint trajectory q of the target robot r (u) Divide the time domain into k segments and scale each joint trajectory with variable time scale; q i,s (t) = q i,r (u), 1≤i≤k; where the variable time scale scaling function is:
[0034]
[0035] Where: q s (t) is the joint trajectory after variable time scale scaling, t is the execution time of the joint trajectory after variable time scale scaling, C i are variable time scale scaling factors, It is a constant in the i-th time period;
[0036] S3. Use the meta-heuristic algorithm to search for n scaling joint trajectories that satisfy the robot motion constraints and time constraints and satisfy the full rank of the matrix A column, where the matrix A is:
[0037]
[0038] in: Wherein: the robot motion constraints include position constraints, speed constraints, and acceleration constraints, that is, the robot's motion must satisfy the position relationship, the specified running speed, and the acceleration; the time constraint is the maximum allowable time for the robot to execute the entire joint trajectory after scaling in step S2; the meta-heuristic algorithm includes but is not limited to the taboo search algorithm, simulated annealing algorithm, genetic algorithm, ant colony optimization algorithm, particle swarm optimization algorithm, artificial fish swarm algorithm, artificial bee colony algorithm, and artificial neural network algorithm. Any one of the above algorithms can be selected, and the above algorithms are all existing algorithms, which will not be described in detail here;
[0039] S4. Control the target robot to execute the n scaling trajectories in order, and collect the energy consumption value E of each scaling trajectory in each time period i,m , where: 1≤i≤k, 1≤m≤n;
[0040] S5. Construct a calculation model for the robot's energy consumption characteristic parameters:
[0041] b=(A T A) -1 A T E, where: b = [b i,1 b i,2 … b i,12 ] T is the energy consumption characteristic parameter of the robot, E=[E i,1 E i,2 … E i,m … E i,n ] T ;
[0042] S6. Construct energy consumption optimization calculation model:
[0043]
[0044] ; Using dynamic programming algorithm, combined with the following time scaling state space expression to scale the parameters Search and iteratively calculate the energy consumption of the joint trajectory based on the energy consumption optimization calculation model and the state transfer equation, and finally obtain the optimal energy consumption scaling trajectory and the optimal energy consumption value that meet the specified conditions:
[0045]
[0046] Where: △ = u i+1 -u i , △ is a constant; the specified conditions are the position constraint, velocity constraint, acceleration constraint of the robot in the specified time period and the execution time of the entire trajectory after scaling;
[0047] In addition, the state transition equation of the dynamic programming algorithm is:
[0048]
[0049] Where: L i represents the minimum joint trajectory energy consumption in the first i-1 time period, E i Indicates that the robot is in state Transfer to The energy consumption of the joint trajectory is calculated by the energy consumption optimization calculation model. The dynamic programming algorithm is a prior art, and its specific process will not be described in detail here.
[0050] Through the above method, the energy consumption of the robot can be optimized and determined through the joint trajectory of the robot and the energy consumption value obtained by executing the joint trajectory, without obtaining other relevant parameters of the robot, thereby effectively improving the efficiency of energy consumption optimization evaluation while ensuring the accuracy of energy consumption calculation.
[0051] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.
Claims
1. A data-driven robot energy consumption optimization method, Features: The following steps are involved: S1. Obtain the preset joint trajectory q of the target robot r (u); where u is the execution time of the preset joint trajectory; S2. Set the preset joint trajectory q of the target robot r (u) Divide the time domain into k segments and scale each joint trajectory with variable time scale; q i,s (t) = q i,r (u), 1≤i≤k; where the variable time scale scaling function is: Where: q s (t) is the joint trajectory after variable time scale scaling, t is the execution time of the joint trajectory after variable time scale scaling, C i are variable time scale scaling factors, It is a constant in the i-th time period; S3. Use the meta-heuristic algorithm to search for n scaling joint trajectories that satisfy the robot motion constraints and time constraints and satisfy the full rank of the matrix A column, where the matrix A is: in: S4. Control the target robot to execute the n scaling trajectories in order, and collect the energy consumption value E of each scaling trajectory in each time period i,m , where: 1≤i≤k, 1≤m≤n; S5. Construct a calculation model for the robot's energy consumption characteristic parameters: b=(A T A) -1 A T E, where: b = [b i,1 b i,2 …b i,12 ] T is the energy consumption characteristic parameter of the robot, E=[E i,1 E i,2 …E i,m …E i,n ] T ; S6. Construct energy consumption optimization calculation model: ; The dynamic programming algorithm is used to combine the following time-scaling state space expression to scale the parameters Search and iteratively calculate the energy consumption of the joint trajectory based on the energy consumption optimization calculation model and the state transfer equation, and finally obtain the optimal energy consumption scaling trajectory and the optimal energy consumption value that meet the specified conditions: Among them: Δ=u i+1 -u i , Δ is a constant; the specified conditions are the position constraint, velocity constraint, acceleration constraint of the robot in the specified time period and the execution time of the entire trajectory after scaling; In addition, the state transition equation of the dynamic programming algorithm is: Where: L i represents the minimum joint trajectory energy consumption in the first i-1 time period, E i Indicates that the robot is in state Transfer to Energy consumption of joint trajectories calculated by the energy consumption optimization calculation model.
2. According to the data-driven robot energy consumption optimization method of claim 1, Features: The robot motion constraints include position constraints, speed constraints and acceleration constraints.
3. The data-driven robot energy consumption optimization method according to claim 1, Features: The time constraint is the maximum allowed time for the robot to execute the entire joint trajectory after scaling in step S2.
4. According to the data-driven robot energy consumption optimization method of claim 1, Features: Meta-heuristic algorithms include but are not limited to taboo search algorithm, simulated annealing algorithm, genetic algorithm, ant colony optimization algorithm, particle swarm optimization algorithm, artificial fish swarm algorithm, artificial bee colony algorithm, and artificial neural network algorithm.
