CN100483427C - Computer aided cue and ball selecting system and method therefor - Google Patents
Computer aided cue and ball selecting system and method therefor Download PDFInfo
- Publication number
- CN100483427C CN100483427C CNB2005101376889A CN200510137688A CN100483427C CN 100483427 C CN100483427 C CN 100483427C CN B2005101376889 A CNB2005101376889 A CN B2005101376889A CN 200510137688 A CN200510137688 A CN 200510137688A CN 100483427 C CN100483427 C CN 100483427C
- Authority
- CN
- China
- Prior art keywords
- club
- ball
- parameter
- flight condition
- hitting
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Abstract
The invention relates to a computer-aided selecting ball arm and ball system and method, it mainly builds a database with the limb property of user, parameters of the ball arm and ball, movement, and the flight status, the database is used to train a decision-making module to choose a proper ball arm, the parameters of the ball arm are then input into the simulation module to be analyzed, when it fulfills the target condition of user, it can be output as the optimal solution, it can achieve the effects of saving manpower, time and money.
Description
Technical field
The present invention relates to a kind of system and method thereof of selecting club and ball, be in particular a kind of machine learning (machine learning) algorithm of carrying out on computers, reach the system and the method thereof of accurate selection club and ball to carry out the mode of emulation.
Background technology
Everyone is inequality in the posture that swings, strength, the speed of golf, and therefore, everyone is the also difference to some extent of club of suitable use.Traditional selection club method can only be at height, body weight, pushing ... the equidimension condition is done simple the selection, and more the practice of distribution also can be measured the reference that the speed that swings is used as selecting bar.But after selecting club, also needing to carry out the actual sets bar beats for professional sportsman's examination again, because professional sportsman more can accurately shoot suitable power and angle compared with general user, the result who waves according to examination is benchmark, experience by trial and error pricing (try and error) or configuration person, or existing form finds out the club parameter that is fit to this user, whether is fit to this user with decision.
If not best of breed, configuration person changes a parameter condition to select second group of club parameter according to linear relationship again, after beating and revise through repeatedly group bar, examination then, can obtain real this user's of being fit to club.Select the bar method yet this kind is traditional, owing to select the club parameter entirely with the personal experience, standard also is not easy control, and can't consider the relevance between multiple parameter; Moreover, select club of institute or ball, be that requirement is reached in just that test of hitting several times probably, and the condition that can't guarantee the user occurs still reaching requirement after some change, and must be through repeatedly trying to wave, organize bar ... Deng action, very labor intensive, material resources and time obviously are not efficient methods.
But, continuous progress along with science and technology, it is more than ever before to measure the characteristic that swings now, for example: bulb speed, the ball speed that ball is produced by the back of hitting, the path swings, face angle ... or the like, in order to utilize these abundanter parameters, select the shortcoming of bar to improve tradition, at No. the 6192323rd, United States Patent (USP) disclosed " Method for matching golfers with a driver and ball ", and at No. the 6083123rd, United States Patent (USP) disclosed " Method for fitting golf clubs for golfers ", be to change by computer program and carry out choose this step of club parameter by configuration person's experience or existing form, method improvement although it is so traditional bar method of selecting, but at the group bar, examination is beaten ... still need to be undertaken etc. step by professional sportsman, and on the other hand, by the selected optimum value of these patents, no matter be to use the form or the method for linear proximity (curve fitting), its result and real conditions all have one section sizable gap.
Therefore, how to find out a kind of method that can effectively select club or ball, the feasible more suitable user of club who selects is an important problem in fact.
Summary of the invention
In view of above problem, fundamental purpose of the present invention is to provide the system and the method thereof of a kind of Computer Aided Selection club and ball, utilize the parameter of limbs characteristic, club and the ball of using the user, adding hit the action parameter and hit after flight condition set up a database that is stored in memory module, after carrying out simulation analysis by computing machine, not only can more accurate selection club, and save group bar, examination money and the time cost beaten, reach that real selection speed is more quick, the result is also comparatively accurate.
Therefore, for reaching above-mentioned purpose, the method for a kind of Computer Aided Selection club disclosed in this invention and ball must comprise the following step:
At first, must in memory module, set up club and the physical parameter of ball and the database of this club flight condition of ball under the different situations of hitting respectively in advance, come again just can database all kinds of parameter data set up an optimized algorithm model, and according to the ball flight condition objective function of received setting, the club parameter of one club, one user uses a group of this club with the upper limb body characteristic, the parameter of flight condition of this ball was selected another club parameter of memory module after the parameter of action of hitting more than a group and many hit, then, with Computer Simulation mode this club according to input, this user uses a group of this club with the upper limb body characteristic, the parameter of this action of hitting and hit after the flight condition emulation of this ball hit this club that decision-making module chooses to obtain the flight condition of a ball, do you and judge that this flight condition meets the objective function of the ball flight condition of setting? when the ball flight condition meets the condition of accuracy that the user sets or long distance, then can be with this club or the output of the parameter of ball, if be not inconsistent with the condition of setting, then to once select the bar action again, can meet club or the ball that the user imposes a condition up to choosing.
About feature of the present invention and embodiment, conjunction with figs. and most preferred embodiment are described in detail as follows.
Description of drawings
Fig. 1 is a system architecture diagram of the present invention;
Fig. 2 is a method flow diagram of the present invention;
Fig. 3 is the method flow diagram that the present invention utilizes genetic algorithm to reach precisely to choose shuttles bar or ball; And
Fig. 4 is the organigram of Connectionist model of the present invention.
The primary clustering symbol description:
40 Connectionist models
110 load modules
120 memory modules
130 decision-making modules
140 emulation modules
150 correcting modules
160 output modules
410 input layers
420 hide layer
430 output layers
440 neural units
Objective function, the club parameter of a club, a user of the ball flight condition that step 210 receive to be set use a group of this club hit with the upper limb body characteristic, more than one group the parameter of action and complex hit after the flight condition of this ball
The club that step 230 calculates this objective function in the Computer Simulation mode is waved the flight condition behind the ball
Does step 250 judge that the flight condition of ball meets the requirements?
The optimal club parameter of step 270 output
The parameter (Initial Population) of the initial group of step 310 input
Does step 370 judge that the club parameter value can reach requirement?
Embodiment
The present invention will disclose a kind of Computer Aided Selection club and ball system and method thereof.In below of the present invention, describing in detail, will multiple specific details be described so that complete description of the present invention is provided.Yet, for those of ordinary skills, and can not need to use these specific detail just can implement the present invention, perhaps can be by utilizing the assembly or the method that substitute to implement the present invention.Under other situation, do not explain known method, program, parts and circuit especially, in order to avoid unnecessarily obscure emphasis of the present invention.
Please refer to Fig. 1, it is a system architecture diagram of the present invention, reach and utilize computing machine to carry out the action of assisted Selection club or ball, must be earlier with all supplemental characteristic digitizings, make computing machine can from data, carry out the step of statistical study, wherein these parameters have comprised human parameters, or perhaps limbs characterisitic parameter, for example: height, body weight, age, sex, the race, the length of playing a ball game and wrist are to floor level ... etc. data, the another kind of parameter relevant with the people is meant people's behavior parameter, for example: the velocity distribution when swinging, the time point that wrist discharges, wrist is returned positive angle, bar head speed, upper boom height and batting (inside out) in the score inboard ... Deng numerical value, the parameter relevant with club is for example following several: the DE of shaft (flex), shaft torsion (torque), maximum camber (kick point), shaft length (shaft length), bar handle angle (lie angle), close angle (close angle), value of moving behind the pole face (offset), C.G., moment of inertia (MOI), club is nose heave (head weight), pole face spring constant (COR) ... Deng.
Please refer to Fig. 2, it is a method flow diagram of the present invention, above-mentioned these data can be created as a database (step 210) prior to memory module 120 according to various data categories in advance in advance, and also comprise the flight condition parameter of ball after hitting in this database, for example: bulb speed, ball speed (ball speed), rotating speed, spherical angle, position angle (azimuth angle), emergence angle (launch angle), (back spin) turns round, sidespin (side spin) and bulb behavior (dynamic loft, dynamic lie, and dynamic openface) ... Deng.
After setting up the database that comprises above-mentioned data, just can utilize these data to train a decision-making module 130, resulting action parameter in the time of can swinging by this decision-making module 130 and according to load module 110 user's that receives limbs characteristic and utilization after finishing, select the one group of optimal club or the parameter (step 220) of ball, this decision-making module 130 is an optimized algorithm model normally, for example: greedy algorithm (Greedy algorithm), decision tree (Decision Tree) learning method, ant algorithm (ant colony algorithm), neural network (neural network), reaction surface method (RSM, Response Surface Methodology) and genetic algorithm (GA, Genetic Algorithm) ... Deng, wherein the most often be used to solve optimized problem with first kind, but it is a kind of mode of dealing with problems of intuition, only find out present optimum solution during each calculating, but final result but may not be the optimum solution in the whole solution space (solution space), just is easy to fall into local optimum and separates (local optimalsolution).
For avoiding taking place such phenomenon, the inventor selects to use genetic algorithm to be used as decision-making module 130, genetic algorithm is mainly with reference to the natural evolution rule of " survival of the fittest in natural selection, the survival of the fittest " in the Darwinian evolution, utilize the biological process of evolving naturally of emulation, set up " artificial gene system " (an Artificial Genetic System) who possesses natural characteristic, again this process is incorporated the process of problem solving.Adopt this algorithm benefit not only can avoid falling into the trap of suboptimization, and can obtain universe optimum solution (globaloptimal solution), and when computing, only need set the objective function (Objectivefunction) of requirement, do not need other supplementary (as differential, the continuity of function), and be not subjected to the restriction of continuous parameters, so be fit to the objective function of all kinds of problems.
Please refer to 3 figure, this is the method flow diagram that utilizes genetic algorithm to reach precisely to choose shuttles bar or ball, supposes to select club, and then the parameter of ball needs to carry out after earlier given again; At first, in the step of the parameter (Initial Population) (step 310) of importing initial group, be to become binary chromosome coding with the club Parameters Transformation, suppose that chromosome length is 8, and group's size is 10, promptly takes out 10 group chromosomes at random as parent from the mating pond, is exemplified as x1=10011011, x2=01001101, x3=00111010, x4=11110101, x5=11101000, x6=00000010, x7=10101101, x8=00001011, x9=10101000, x10=11110110, wherein the round values of the coding representative shown in x2 is 64+8+4+1=77.
Next just can require target according to what the user set, for example: the distance of being flown in the back of hitting, or the accuracy of hitting decides one group of objective function (perhaps be called and adapt to function (fitnessfunction)) f (x), corresponding target function value (step 320) is calculated in this 10 groups of club parameters substitution, suppose that the user promptly is set at the beginning and only select the highest chromosome of 2 groups of adaptive values (fitness value) in the group and carry out the genetic manipulation program, and 2 groups minimum (step 330) of superseded adaptive value.
The user can set mating rate, mutation rate ... etc. the probability parameter, with emulation gene evolution process, make and finally can produce the strongest species of adaptive faculty, optimum solution just, wherein the evolutionary process in the genetic algorithm has three key steps: in the step of gene duplication (reproduction) (step 340), allow the feature of parental generation remain into filial generation, in order to avoid along with the evolution of time, the strong point that organic evolution is come out has been vanished from sight on the contrary.Adaptedness according to each species decides among the next generation and should be eliminated, or duplicates and what a kind of calculating process of the number that keeps.
Reproduction process has two kinds of forms: (a) Wheel-type is selected (roulette wheel selection) and (b) competitive mode selection (tournament selection), wherein (a) is in the evolutionary process in each generation, the size of at first complying with the adaptive value of each species (chromosome) is cut apart the position of wheel disc, the words that adaptive value is big more, the area ratio that then occupies on wheel disc is also big more, the selected probability in the mating pond of the big more representative of the area ratio that each species occupies on wheel disc is big more, then the picked at random wheel disc a bit, its pairing species promptly are selected in the mating pond.And (b) promptly be in the evolutionary process in each generation, to choose at first randomly in two or more species (chromosome), the species with maximum adaptation value are selected delivering in the mating pond.
In the step of gene mating (crossover) (step 350), it promptly is the mating expression formula, from a group, select any two chromosomes randomly according to the mating rate that had before set, through some gene information of chiasmatypy each other and produce two new chromosomal a kind of processes, can allow chromosome intercourse Useful Information by this step, so that chromosome obtains higher adaptive value, to improve filial generation.Generally speaking the mating computing can be used: (a) single-point mating; (b) two point mating; And (c) evenly mating; Lifting a single-point mating is example: suppose in aforesaid 10 group chromosomes, x1 and x2 can obtain the highest adaptive value, wherein after the 5th (bit), begin to carry out mating, then can obtain two groups of child chromosome of x11=10011101 and x12=01001011, just may create the species of higher adaptive value thus.
And the 3rd step is to suddenly change (mutation) (step 360), and the process of sudden change is the chromosome of choosing at random, and the picked at random catastrophe point, and then controlled by mutation rate by the gene information mutation process generation probability that changes in the chromosome.Mutation process can be exactly 0 of character string to be become 1,1 become 0 for binary character string at term single gene or to the suddenly change calculation or in word cover sudden change mode for it of a plurality of genes.Two filial generations of same continuity step 350 gained are the example explanation, the position is the 1st if x11 gets sudden change at random, that is to say that new x11 will become 00011101 after suddenling change, if x12 also is assigned at random and need be suddenlyd change, and catastrophe point is at the 6th, and then new x12 will become 01001111.Can make genetic algorithm add new parameter value by this step, and can not miss some Useful Informations, and the probability of sudden change basically is very low that the words of Ruo Taigao may cause Useful Information to lose on the contrary.
In these club parameter value substitution objective functions, judge whether the highest club parameter of adaptive value meets accuracy or flying distance (step 370) that the user sets, if then be not the evolutionary process of continuity above-mentioned steps 320 to step 370, superseded unaccommodated chromosome progressively, actual in the process that develops, the number of group may be thousands of or more, and mortality may be more than 0.5, therefore the chromosome in each generation will be very big in the change gap of gene, even if sudden change failure or mating failure, because parent still exists, so still keeping the high gene of adaptive value, so after through evolution repeatedly, just may restrain and reach an optimum solution, perhaps the user also can set the number of times of evolution, to avoid still can stopping the calculating of computing machine under convergent situation not and export preferable separating.
When this one preferable separate or the club parameter of optimum solution by decision-making module 130 outputs after, just then carry out simulation analysis with this parameter by emulation module 140, calculate this club hit the back ball flight condition (step 230), with computer-aided engineering (CAE, Computer Aided Engineering) is example, can calculates ball flight result by wired ultimate analysis, the course that the swings input that measures, just can calculate the process of club impact bead, and obtain the behavior that flies out of ball.
The another kind of flight condition of calculating cross hit is to use the mode of neural network to calculate, please refer to Fig. 4, this is the organigram of a Connectionist model 40, it is input more than, many outputs and signal forward transmit the high-speed calculating unit of (feed-forward), it forms (and the circle part among the figure) by 440 of many interconnected neural units (neuron), each neural unit 440 comprises one and adds up contact (summing junction), to be added up earlier row totalling again from the signal of other neural unit 440 individually, and all neural units 440 adhere to three layers separately: input layer (input layer) 410, hide layer (hidden layer) 420 and output layer (output layer) 430.
Each output signal by input layer 410 neural units 440 needs after different weighting (weight), the different neural units 440 of layer 420 are hidden in feed-in again, and each also needs after different weightings the different neural units 440 of feed-in output layer 430 again by hiding output signal that layer 420 neural unit 440 send.Utilize club parameter in the database and limbs characterisitic parameter as input value, the offline mode parameter of ball is as output valve in the database, train a Connectionist model 40 (for a kind of regression model), this model must be trained and input value in the database and output valve can be made an accurate correspondence, and after corresponding result all is tending towards convergence, just the behavior parameter that swings that measures in the time of can utilizing the user of input to swing, and the limbs characterisitic parameter estimates the behavior that flies out of ball in this Connectionist model 40, and can calculate ball flight condition (step 240).
Because decision-making module 130 is different algorithms with emulation module 140 employings, therefore, the club parameter that decision-making module 130 is picked out, after process is carried out emulation testing, judge whether the ball flight condition can meet the user and set the accuracy of hitting, perhaps flying distance fall short of (step 250), if can't satisfy user's requirement condition after the emulation, export a corrected signal to decision-making module 130 (step 260) by correcting module 150 again, what decision-making module 130 just began to carry out second leg selects the bar running, the club parameter remains at random and produces, select with genetic algorithm, and, then this is reached the club parameter output of requirement result by output module 160 if decision-making module 130 selected club parameters meet the requirements in the result of emulation really.Same, if will select a suitable ball, then be after the club parameter is selected earlier, to select according to above-mentioned step.
Thus, just can reach to use a computer easily by the present invention and carry out simulation analysis, not only can more accurate selection club and ball, and save money and the time cost that group bar, examination are beaten, reach that real selection speed is more quick, the result is also comparatively accurate.
Though the present invention with aforesaid preferred embodiment openly as above; right its is not in order to qualification the present invention, any those of ordinary skills, without departing from the spirit and scope of the present invention; can do some changes and revise, therefore scope of patent protection of the present invention is as the criterion with claim.
Claims (16)
1, a kind of system of Computer Aided Selection club is characterized in that, this system has:
One memory module is in order to the physical parameter that stores many clubs and ball and many flight condition of the ball of this club under the different situations of hitting respectively;
One load module, in order to objective function, the club parameter of a club, a user of ball flight condition that receive to set use a group of this club hit with the upper limb body characteristic, more than one group the parameter of action and many hit after the flight condition of this ball;
One decision-making module uses the flight condition of one group of this club this ball with the parameter of upper limb body characteristic, this action of hitting and after hitting in order to this club parameter, this user according to input, chooses another club parameter of memory module;
One emulation module, use the flight condition of one group of this club this ball with the parameter of upper limb body characteristic, this action of hitting and after hitting in order to this club parameter, this user according to input, emulation is hit this club that decision-making module chooses obtaining the flight condition of a ball, and judges whether this state that flies out meets the objective function of the ball flight condition of setting; And
One output module is in order to the club of the objective function of exporting this ball flight condition that meets setting.
2, system according to claim 1 is characterized in that, this decision-making module is reached by genetic algorithm, and this genetic algorithm comprises step:
Step a1, each target function value of calculating ball flight condition;
Step b1 becomes the scale-of-two chromosome coding with this club Parameters Transformation, selects the highest chromosome of adaptive value;
Step c1 duplicates gene;
Steps d 1 is carried out mating to gene;
Step e1 suddenlys change to gene;
Step f1 in club parameter value substitution objective function, judges whether the highest club parameter value of adaptive value can reach requirement;
Wherein, in step f1, if, then finish, if not, then continue step a1 to step f1.
3, system according to claim 1 is characterized in that, this emulation module is to be a computer-aided engineering, calculate ball flight result by wired ultimate analysis, the course that swings input with measuring calculates the process of club impact bead, and obtains the behavior that flies out of ball.
4, system according to claim 1, it is characterized in that, this emulation module is to be a Connectionist model, be input more than, export and high-speed calculating unit that signal forward transmits more, it is made up of the interconnected neural unit of majority, each neural unit comprises one and adds up contact, is the signal from other neural unit is added up earlier row totalling more individually, and all neural units adhere to three layers separately: input layer, hiding layer and output layer.
5, system according to claim 1 is characterized in that, the flight condition of this ball is that a bulb speed, a ball speed, a rotating speed, a spherical angle, a position angle, an emergence angle, turn round, a sidespin and a bulb behavior.
6, system according to claim 1 is characterized in that, the objective function of the ball flight condition of this setting is the distance for a flight.
7, system according to claim 1 is characterized in that, the objective function of the ball flight condition of this setting is the accuracy of hitting for.
8, system according to claim 1 is characterized in that, this system also comprises a correcting module, exports a corrected signal to this decision-making module in order to judge not meet in this club parameter when this setting requires.
9, system according to claim 8 is characterized in that, this corrected signal is in order to cause this decision-making module to carry out time time another club parameter of selection.
10, a kind of method of Computer Aided Selection club, be applied in the physical parameter that stores many clubs and ball and many respectively this club it is characterized in that on a memory module of the flight condition of the ball under the different situations of hitting this method comprises the following step:
(a) objective function, the club parameter of a club, a user of the ball flight condition of receive setting use a group of this club hit with the upper limb body characteristic, more than one group the parameter of action and many hit after the flight condition of this ball;
(b) use the flight condition of one group of this club this ball with the parameter of upper limb body characteristic, this action of hitting and after hitting according to this club parameter, this user of input, choose another club parameter of memory module;
(c) use the flight condition of one group of this club this ball with the parameter of upper limb body characteristic, this action of hitting and after hitting according to this club parameter, this user of input, emulation is hit this club that decision-making module chooses to obtain the flight condition of a ball;
(d) judge whether this state that flies out meets the objective function of the ball flight condition of setting; And
(e) export the club parameter of the objective function of this ball flight condition that meets setting.
11, method according to claim 10 is characterized in that, should (d) step if judge and do not meet, then repeat again (b) step to (d) step to select another club parameter.
12, method according to claim 10 is characterized in that, should (b) step be to choose the action that this is chosen with genetic algorithm, and this genetic algorithm comprises step:
Step a1, each target function value of calculating ball flight condition;
Step b1 becomes the scale-of-two chromosome coding with this club Parameters Transformation, selects the highest chromosome of adaptive value;
Step c1 duplicates gene;
Steps d 1 is carried out mating to gene;
Step e1 suddenlys change to gene;
Step f1 in club parameter value substitution objective function, judges whether the highest club parameter value of adaptive value can reach requirement;
Wherein, in step f1, if, then finish, if not, then continue step a1 to step f1.
13, method according to claim 10 is characterized in that, should (c) step reach with a computer-aided engineering, calculate ball flight result by wired ultimate analysis, the course that swings input with measuring calculates the process of club impact bead, and obtains the behavior that flies out of ball.
14, method according to claim 10, it is characterized in that, should (c) step reach with a Connectionist model, this Connectionist model is input more than, the high-speed calculating unit that many outputs and signal forward transmit, it is made up of the interconnected neural unit of majority, each neural unit comprises one and adds up contact, to be added up earlier row totalling again from the signal of other neural unit individually, all neural units adhere to three layers separately: input layer, hide layer and output layer, each by the output signal of input layer neural unit after different weightings, the different neural units of layer are hidden in feed-in again, and each is by hiding the also different neural units of feed-in output layer again after different weightings of output signal that layer neural unit send, utilize club parameter in the database and limbs characterisitic parameter as input value, the offline mode parameter of ball is as output valve in the database, input value in the database and output valve are made an accurate correspondence, and after corresponding result is tending towards convergence, the behavior parameter that swings that measures when utilizing the user who imports to swing, and the limbs characterisitic parameter estimates the behavior that flies out of ball in this Connectionist model, calculates the ball flight condition.
15, method according to claim 10 is characterized in that, the objective function of this ball flight condition is the distance of a flight.
16, method according to claim 10 is characterized in that, the objective function of this ball flight condition is the accuracy of hitting.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005101376889A CN100483427C (en) | 2005-12-31 | 2005-12-31 | Computer aided cue and ball selecting system and method therefor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005101376889A CN100483427C (en) | 2005-12-31 | 2005-12-31 | Computer aided cue and ball selecting system and method therefor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1991848A CN1991848A (en) | 2007-07-04 |
| CN100483427C true CN100483427C (en) | 2009-04-29 |
Family
ID=38214101
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2005101376889A Active CN100483427C (en) | 2005-12-31 | 2005-12-31 | Computer aided cue and ball selecting system and method therefor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100483427C (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101398778B1 (en) * | 2012-03-30 | 2014-05-27 | 스미토모 고무 고교 가부시키가이샤 | Golf club shaft fitting method |
| TW201415272A (en) * | 2012-10-12 | 2014-04-16 | Ind Tech Res Inst | Method for swing result deduction and posture correction and the apparatus of the same |
| JP2016209228A (en) * | 2015-05-07 | 2016-12-15 | セイコーエプソン株式会社 | Swing analyzer, swing analysis method, swing analysis program and swing analysis system provided with swing analyzer |
-
2005
- 2005-12-31 CN CNB2005101376889A patent/CN100483427C/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN1991848A (en) | 2007-07-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Arabzad et al. | Football match results prediction using artificial neural networks; the case of Iran Pro League | |
| Haiyun et al. | Sports performance prediction model based on integrated learning algorithm and cloud computing Hadoop platform | |
| CN109499068A (en) | The control method and device of object, storage medium, electronic device | |
| Samsuden et al. | A review paper on implementing reinforcement learning technique in optimising games performance | |
| Oughali et al. | Analysis of NBA players and shot prediction using random forest and XGBoost models | |
| CN100483427C (en) | Computer aided cue and ball selecting system and method therefor | |
| Babaee Khobdeh et al. | Clustering of basketball players using self-organizing map neural networks | |
| CN111330255B (en) | Amazon chess-calling generation method based on deep convolutional neural network | |
| Ying et al. | A novel optimization algorithm for BP neural network based on RS-MEA | |
| Omkar et al. | Cricket team selection using genetic algorithm | |
| Nakashima et al. | Off-line learning of soccer formations from game logs | |
| Du et al. | Uprising E-sports Industry: Machine learning/AI improve in-game performance using deep reinforcement learning | |
| Muthuswamy et al. | Bowler performance prediction for one-day international cricket using neural networks | |
| Zhao et al. | Tracing mechanism of sports competition pressure based on backpropagation neural network | |
| CN112131519A (en) | Application method of improved volleyball super league algorithm in knapsack problem | |
| Cao et al. | Research on the DouDiZhu's playing strategy based on XGBoost | |
| Wang | Parameter Optimization and State Evaluation of Basketball Teaching Based on BPNN | |
| Wang et al. | The design of sports games under the internet of things fitness by deep reinforcement learning | |
| Park et al. | Deep Q-learning using redundant outputs in visual doom | |
| Yuan et al. | Recommending Bids on Dou-DiZhu Poker Games: A Deep Learning Approach | |
| Guan et al. | Special Physical Characteristic and Training Strategy of Badminton Based on Machine Learning Algorithm | |
| Bytyçi et al. | Combination of Genetic Algorithm and Neural Network for Independent Game Playing | |
| Qu | Prediction and Analysis of the Physical Test Scores Based on BP Neural Network and Principal Component Analysis Algorithm | |
| Scopinho | Artificial neural networks on recruiting athletes | |
| Huang et al. | Quantitative of badminton practice based on wireless sensor network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |