CN102346815A - Digital biological system for simulating biological competition and evolution process - Google Patents
Digital biological system for simulating biological competition and evolution process Download PDFInfo
- Publication number
- CN102346815A CN102346815A CN2011102888288A CN201110288828A CN102346815A CN 102346815 A CN102346815 A CN 102346815A CN 2011102888288 A CN2011102888288 A CN 2011102888288A CN 201110288828 A CN201110288828 A CN 201110288828A CN 102346815 A CN102346815 A CN 102346815A
- Authority
- CN
- China
- Prior art keywords
- biological
- digital
- numeral
- offspring
- competition
- 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.)
- Granted
Links
Images
Abstract
The invention relates to a digital biological system for simulating a biological competition and evolution process. According to one aspect of the invention, the process from birth, reproduction, variation, competition among individuals to death of a biological individual is simulated by programs, and each single program represents a biological individual. Once a parent program is started, the duplicate child programs can be initialized and die automatically; as the individuals are different in the execution times for implementing a certain loop command before child duplication and the CPU (central processing unit) processing time to be consumed, different competitive advantage degrees are formed among different programs, and the numbers of the reproduced child programs are different as well. By setting different parameters, the programs can simulate different variant children and simulate the competition as well as the evolution process of selecting the superior and eliminating the inferior of biological individuals, thereby providing a new experiment tool of a bionic model.
Description
Technical field
The invention belongs to biological (digital organism) field of numeral of artificial life.Specifically; The present invention relates to the bionical instrument of a kind of computing machine; The independently initialization of program of the digital bion of representative in this bionical instrument, can the replicon program, have between the different subroutine of certain probability generation, the subroutine and have different competitive power, and each subroutine has certain life-span.The bionical instrument of this computing machine can be applied to simulate the bionical experiment of biological competition and evolutionary process.
Background technology
Along with 20th century computer science and the development of bio-science, a new interdisciplinary study---artificial life has been born in the field, intersection forward position of bio-science and computer science.Artificial life is meant the manual system of copying biological vital movement rule, Self-organization Evolution system and setting up, and it is through to the abstract of various life forms and mechanics and conclusion, and attempting to construct a kind of is the life form of carrier with the non-biological material.Because the artificial life platform has the characteristic than the more convenient operation of natural system, be more convenient for repetition and accurate control, so it is learning object with the bion on the one hand; After setting up the artificial life platform on the other hand; Can carry out dynamic experiment from a plurality of different levels and angle simulation life system; Come test organisms learn in the experimental result of various theory hypothesis and institute's established model; Help to understand natural life movable essential characteristic and evolution rule, for modern biology research provides new method and new tool.Artificial life is also made a general reference various man-made systems with natural life system action characteristic; It carries out abstract to characteristics such as self-organization in the natural life system, self-reproduction, metabolism, study and evolution and concludes; And, explore the mechanism of vital movement with this with these dynamic processes of computer simulation.
The numeral biology is an important branch in the artificial life field; It is simulated object with the biological evolution; With program coding itself as single biology, a series of behavioral activities such as can independently survive, vie each other between self-procreation, variation, death and individuality with biological characteristic.Because digital biology has the characteristic with immediate foundation characteristic of true life phenomenon and self-organization evolution system; It begins in being born; Just become one of most active research focus in the artificial life field, and be applied in a plurality of biological study directions such as selection effect under biogenesis and evolutionary process, genome mutation, environmental pressure.
The biological design philosophy of numeral originates from the Corwars recreation of AT&T Labs's sixties in last century, the Tierra system that nineteen ninety Thomas Ray has write, the formal birth of declaring digital biosystem.Henceforth, except the subsequent improved version of Tierra system, relevant scholar also further develops a plurality of digital biosystems such as Avida, Amoeba, Echo.These systems have all accomplished and can simulate biological breeding and mutation process with independently thread or process.And with program done experiment explanation some about biological phenomenons such as relation, neutral evolutionary process, social organization's structure between bio-diversity and environmental complexity principles behind.
Yet the digital biosystem in past does not have the natural race problem between the fine solution individuality; The main mode that relies on the answer combination points-scoring system of a series of functions of doing mutually between some unit programs of game rule exclusive disjunction; Come the competition difference between the generating routine; And need distribute to the different operational system resource of each program (distributing the time) by artificial pre-defined rule like cpu, reward the selection course of simulating nature with these corresponding resources.This competition that can cause between the individual program is not real free competition; Artificial preset selection rule or points-scoring system will directly influence these digital biological programs of experimental result and can not independently accomplish whole life; Must rely on certain resource external to distribute or instruction could complete all life cycles; Simultaneously digital biosystem neither open system, must in virtual computer environment, could carry out.These shortcomings have all limited further developing and using of digital biosystem.
Summary of the invention
The invention provides a kind of digital biosystem that can simulate biological free competition process and evolutionary process.Vital movements such as the birth of each computer program in the system (digital bion) simulation bion, variation, existence, breeding; Each computer program is accomplished the computer hardware resources such as cpu of the inequality of offspring's the required consumption of life cycle from be born to duplicating in the system, the environmental resource of required consumption when having represented different phenotype bions to accomplish life cycle separately.Like this, the difference owing to the computer resource utilization ratio makes to have produced direct competitive power difference between the computer program, can represent the competition difference between the biological different phenotypes of being simulated with it.Use the method; Biological competition and the selection course of feasible numeral; No longer need artificial game rule or the points-scoring system formulated in advance in the external digital biosystem; Therefore the biological competition process of numeral receives the influence of artificial subjective factor less in the native system, and experimentation and experimental result also more approach biological competition process and the result under the state of nature.The present invention has also broken away from the past to be needed artificially to set fitness function or select this link of coefficient in the bionical mathematical modeling when simulation competition link, and can in experiment, reflect the dynamic evolutionary process of digital biotic population.
Description of drawings
Thread (Worm) in the digital according to an embodiment of the invention biosystem of Fig. 1 illustration.
The operational scheme of the thread (Worm) in the digital according to an embodiment of the invention biosystem of Fig. 2 illustration.
Fig. 3 is illustrated in and continues to occur in the evolutionary process of favorable variation, and the activity of various phenotype types changes.
Fig. 4 illustration is according to embodiments of the invention, with the experimental result of the evolutionary process of digital biosystem simulation drawing 3.
Fig. 5 is illustrated in and continues to occur in the evolutionary process of neutral variation, and the activity of various phenotype types changes.
Fig. 6 illustration is according to embodiments of the invention, with the experimental result of the evolutionary process of digital biosystem simulation drawing 5.
When the intermediate form in Fig. 7 illustration evolution path was all unfavorable variation, the activity of various phenotype types changed.
Fig. 8 illustration is according to embodiments of the invention, with the experimental result of the evolutionary process of digital biosystem simulation drawing 7.
Fig. 9 illustration continues to occur in the evolution path of unfavorable variation, and when having an advantage type, the activity of various phenotype types changes.
Figure 10 illustration is according to embodiments of the invention, with the experimental result of the evolutionary process of digital biosystem simulation drawing 9.
Embodiment
Use following process and principle in the digital biosystem program design of the present invention:
(1) program has the ability of duplicating the offspring, and can produce n offspring's subroutine, and each subroutine all has thread or the process of oneself.(n is the integer more than or equal to 0)
(2) offspring's program of program copy, autonomous initialization.
(3) program inside can provide certain variation ability, when duplicating the offspring, has certain probability can produce different offspring's subroutines.
(4) program has certain life-span, and when externally system's available hardware resource is lower than certain condition, or the life time point of program is when arriving, and program can stop operation automatically, withdraws from execution.
(5) after program withdraws from, must discharge shared all hardware resource.Need memory resources release at once before program withdraws from, withdraw from the back cpu resource and discharge automatically.
(6) program is with the state of the hardware resource environment of the dynamic mapping prerequisite as breeding.
(7) program has the absolute construction of relative closure, in case after the operation, do not need external command just can continue to carry out.
The basic operational scheme of the biological program of numeral is following:
Each numeral is biological to be a thread in native system, and we also can be referred to as a Worm.Its command sequence to the effect that of a Worm, the actual execution content of Worm is by its command sequence decision.An order is exactly a bit of code, such as internal memory of application, does a bit of circulation or the like.Just according to the content of command sequence, carry out by order ground of an order after starting for Worm.In addition, also to write down some some other information of start-up time or the like.The basic structure of Worm is as shown in Figure 1.
A Worm at first notes the start-up time of oneself after being activated.Fill order is if inadequate resource then finishes automatically.Run into checkpoint (check point) after each fill order finishes, all can look over " current time-start-up time " and whether surpass the predetermined time limit (life-span), if overtime, Worm will finish immediately, and can not produce the next generation.If do not have overtimely to judge that then whether cpu busy percentage surpasses threshold value, if surpass threshold value be cpu idle time seldom then process will be suspended a period of time and then entering checkpoint.If all normal, then judge in the command sequence whether also have other orders, if any then carrying out Next Command.As if there not being other orders, check then whether current condition satisfies the generation filial generation, also finish if satisfy then produce filial generation, otherwise directly finish.
Worm terminates as " natural death " after producing filial generation, and remaining situation is improper termination, is called Worm " unusual death ".Cause the reason of a Worm unusual death, except overtime, also have inadequate resource.Such as, a Worm is in operational process, and application is less than the memory source of needs, so also with unusual death.
Each Worm will send out a message to watchdog routine when starting and withering away (no matter being normal termination or unusual death).Watch-dog is according to the content of message, operation information that can the real-time update system.The operational scheme of Worm is as shown in Figure 2.
When doing different experiments, can be through the rule and the incompatible realization condition of different of command group of change Worm operation.Through changing this module, under the constant situation of total system structure, can realize different-effect.
System monitor need start before Worm starts in advance.System monitor is independent of total system and exists, and through network system is connected with watchdog routine.Therefore, a plurality of your pupil's attitude system in combination with the multiple computers simulation are a big ecosystem in the future, and the running situation of diverse location all can be got ready to the influence of Worm for environment of observation develops by a watchdog routine unified management.
Exemplary codes
The main modular of code: native system is made up of 4 modules, is respectively Automatic Program evolution module, monitoring module, communication module and accessory module, and they are implemented in the bag separately.
1, Automatic Program evolution module: edu.fudan.jworm2.client
Operation of control thread and the function that develops are implemented in this module all.
Environment: write down some and the relevant parameter of thread operation, such as thread life-span, cpu restriction or the like, these parameters all are effective to all threads.
SystemMonitor: this is the class of supervisory computer system ruuning situation, has only the function of monitoring cpu utilization rate at present.
Worm: thread is the base class of all threads, and various threads (different operational processs, different variation method or the like) can derive from from this thread.
The subclass of RepeatingWorm:Worm is the thread of the operation in present our this system.About the content of this thread, there has been detailed description the front.
StartWormWorld: the effect of this type is the operation that starts thread evolution system, has comprised parameter and has read, work such as the generation of initial thread and startup.
2, system monitoring module: edu.fudan.jworm2.server
Bring in constant renewal in statistical information, and output regularly.
ServerArguments: write down the parameter of some monitoring clients, such as the information output file, sampling time interval of evolution system operation situation or the like.
WormWorldMonitor: monitoring client master routine, its task are exactly constantly to receive the message that various Worm send, and then message are handled.
WormWorldSnapShoter: sampling routine, its task are exactly to read statistical information at set intervals, output to then in the file.
WormWorldInfo: the object that comes to this that the statistical information of evolution system, sampling routine obtain, export according to the content of this object then.
StartWatcher: the monitoring client starter, do parameter and read, work such as initialization communication channel.
3, communication module: edu.fudan.jworm2.communicate
The code of this module is used for monitoring client and the main system end communicates.
Message: the message of monitoring client and master routine end communication.
Messenger: the communicator between monitoring client and the master routine can be used for sending and accepting message.
MessengerServer: monitor a port, have to connect when arriving, generate a Messenger.
4, utility module: edu.fudan.jworm2.util
Some classes all more useful in other three modules have been comprised, such as the parser of parameter, counter.
Exemplary experiment
Among the application, " experiment " this term refers to after given digital biology and environment of living in thereof one or multinomial condition, the process that biological evolution is simulated to numeral.If not specify that " experiment " among the application and the implication of " embodiment " are equal to.Accordingly, " experimenter " among the application is meant the implementer of technical scheme of the present invention.
Although the present invention obviously can be applied to scientific research, the applicant does not hope to limit the instrument that the present invention is only used for scientific research.In fact, it will be understood by those skilled in the art that biopharmaceutical industry, animal husbandry, plant husbandry, microorganism already in the multiple relevant industries, " the evolutionary process simulation that numeral is biological " technology has the application on the industry widely.The application mainly introduces the design of simulation process itself in an embodiment from concise and to the point purpose.The application of biological evolution analogue technique efficiently true to nature on industry is self-explantory, repeats no more at this.
In exemplary experiment, be provided with through the parameters and the rule of configuration file, as its calculation times and the relation that needs the memory headroom size are set whole procedure.Simultaneously, the original state of seed routine is set, like the cycle index of seed routine, initial filial generation number through the another one configuration file.
During operation, whether each digital biology at first Rule of judgment meets, and whether, cpu busy percentage whether enough such as the free memory space is lower than assign thresholds.When all conditions met, the instruction of operation predetermined number of times judged whether overtimely before each circulation beginning, if overtime then digital biology withdraws from, otherwise it is intact up to all instruction operations to continue operation.If this moment is still not overtime, then judge whether to satisfy the follow-on condition of producing, then produce filial generation if satisfy.
Each numeral biology is born and withdraws from Shi Douhui and transmit the state information to system monitor through network, and system monitor is every to output to the information that counts in the specified file at a distance from the set time (as 5 seconds).
Before experiment beginning, the user can to the biological ancestors' individual amount of numeral, each digital biology can reproduction offspring's number, digital biological life cycle, recursion instruction execution number of times (activity when represent individual survival is big or small), mutation rate, the initial values such as plus-minus amplitude of recursion instruction number of times are provided with when making a variation at every turn.After experiment beginning, system is from the biological beginning of several ancestors' numerals, constantly experience birth, variation, breeding, dead process, and individual in population quantity is on the increase, and the dynamic available resources that computer hardware provides will determine the upper limit of population size.Because digital biological program provides inner made a variation ability; Among the offspring who makes digital biological ancestors produce various variation type is arranged; The competitive power of dissimilar digital bions in colony is different, and the individual type that competitive power is dominant can occupy the majority in the colony gradually.Use this process, can be used for simulating the various evolutionary processes of biological variation, breeding, the survival of the fittest.
For example, in experiment, suppose bion from phenotype X, through a, b, c ... wait a series of intermediate phenotypes, constantly accumulation variation arrives Y phenotype type at last.If each variation type in this evolution path all will have competitive edge than previous variation type; Be similar to the gradual change evolutionary process in the Darwinism theory; (height of column amount is the activity of digital bion required completion before raising up seed among the figure shown in the biological mutation process that is experienced of the numeral among Fig. 3; The activity of required completion is big more; Then the system resource of required consumption is many more before the breeding, and the competitive power of this numeral bion is more weak), then the experimental result of evolutionary process is as shown in Figure 4.
For example, in experiment, suppose bion from phenotype X, through a, b, c ... wait a series of intermediate phenotypes, constantly accumulation variation arrives Y phenotype type at last.If the competitive power advantage degree of each variation type in this evolution path is very approaching or identical; Be similar to the evolutionary process of neutral theory; Shown in the biological mutation process that is experienced of the numeral among Fig. 5, then the experimental result of evolutionary process is as shown in Figure 6.
For example, in experiment, suppose bion from phenotype X, through a, b, c ... wait a series of intermediate phenotypes, constantly accumulation variation arrives Y phenotype type at last.If each variation type in this evolution path all is to belong to the more weak bad variation type of competitive power, shown in the biological mutation process that is experienced of the numeral among Fig. 7, then the experimental result of evolutionary process is as shown in Figure 8.
For example, in experiment, suppose bion from phenotype X, through a, b, c ... wait a series of intermediate phenotypes, constantly accumulation variation arrives Y phenotype type at last.If each the variation type majority in this evolution path all is the more weak bad variation type of competitive power; But the d variation type that is positioned in the middle of the path belongs to the favourable type of competition; Shown in the biological mutation process that is experienced of the numeral among Fig. 9, then the experimental result of evolutionary process is shown in figure 10.
Through native system, the dynamic evolutionary process in the time of can simulating biomutation through the individual evolution path of various phenotypes with different competitive power very easily provides one to help the characteristic of research biological evolution and the new method and the new tool of rule.
Claims (6)
1. digital bionic method of simulating biological competition and evolutionary process, its characteristic comprises:
1) in computing environment, create a plurality of numeral biologies, the biological corresponding thread of each numeral wherein provides the recursion instruction that needs cpu to handle in digital biology interior;
2) recursion instruction in the combine digital biology;
3) said digital biology has certain life-span; When the numeral biology is handled recursion instruction in its life-span; Produce n the sub-thread of offspring for this numeral is biological to represent offspring's numeral biological, wherein, the cycle index of the recursion instruction that numeral is biological can change when producing the offspring;
4) when the biological life time point of numeral arrives, withdraw from the biological thread execution of this numeral;
5) the biological repeating step 2 of offspring's numeral that each produced)-4);
Wherein, each digital biological thread is independent of the statistics of system monitor transmission for the biological various states of numeral.
2. like right 1 described digital bionic method, it is characterized in that when the biological generation of numeral offspring subroutine, the cycle index of the recursion instruction of offspring's subroutine morphs with certain probability.
3. like right 1 described digital bionic method, it is characterized in that when n offspring's subroutine of the biological generation of numeral, n is fixed value or random value.
4. like right 1 described digital bionic method; It is characterized in that numeral is biological when being born with death, can be to system monitor independence transmission information; Said system monitor is independent of digital biosystem and exists; With digital biosystem on same main frame, or on different main frames, link to each other through network with digital biosystem.
5. like right 1 described digital bionic method, it is characterized in that said digital bionic method is used to simulate biological breeding, variation, heredity, death process.
6. like right 1 described digital bionic method, it is characterized in that, with the resource of cycle index analog digital bion required consumption before breeding of the biological recursion instruction of numeral.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110288828.8A CN102346815B (en) | 2010-12-28 | 2011-09-26 | Digital biological system for simulating biological competition and evolution process |
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010609783 | 2010-12-28 | ||
| CN201010609783.5 | 2010-12-28 | ||
| CN201110288828.8A CN102346815B (en) | 2010-12-28 | 2011-09-26 | Digital biological system for simulating biological competition and evolution process |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102346815A true CN102346815A (en) | 2012-02-08 |
| CN102346815B CN102346815B (en) | 2014-08-20 |
Family
ID=45545488
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110288828.8A Expired - Fee Related CN102346815B (en) | 2010-12-28 | 2011-09-26 | Digital biological system for simulating biological competition and evolution process |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102346815B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6708141B1 (en) * | 1998-01-16 | 2004-03-16 | The University Of Connecticut | Method for modeling cellular structure and function |
| CN101770560A (en) * | 2008-12-31 | 2010-07-07 | 南方医科大学 | Information processing method and device for simulating biological neuron information processing mechanism |
| CN101952835A (en) * | 2007-09-07 | 2011-01-19 | 克劳利戴维研究公司 | It with the cell the model that the center is carried out system for simulating and method and obtained thus based on cell |
-
2011
- 2011-09-26 CN CN201110288828.8A patent/CN102346815B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6708141B1 (en) * | 1998-01-16 | 2004-03-16 | The University Of Connecticut | Method for modeling cellular structure and function |
| CN101952835A (en) * | 2007-09-07 | 2011-01-19 | 克劳利戴维研究公司 | It with the cell the model that the center is carried out system for simulating and method and obtained thus based on cell |
| CN101770560A (en) * | 2008-12-31 | 2010-07-07 | 南方医科大学 | Information processing method and device for simulating biological neuron information processing mechanism |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102346815B (en) | 2014-08-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Thompson et al. | Food webs: reconciling the structure and function of biodiversity | |
| GB2601643A (en) | Spiking neural network computing system and method for brain-like intelligence and cognitive computing | |
| Ginot et al. | A multi-agents architecture to enhance end-user individual-based modelling | |
| Lorek et al. | Modelling and simulation software to support individual-based ecological modelling | |
| Caplat et al. | Symmetric competition causes population oscillations in an individual-based model of forest dynamics | |
| CN112036540A (en) | Sensor number optimization method based on double-population hybrid artificial bee colony algorithm | |
| Feng et al. | Dynamics of task allocation in social insect colonies: scaling effects of colony size versus work activities | |
| Lebiere et al. | Multi-tasking and cognitive workload in an ACT-R model of a simplified air traffic control task | |
| Corbey | Baldwin effects in early stone tools | |
| Yamauchi et al. | Relaxation of selection, niche construction, and the Baldwin effect in language evolution | |
| Herajy et al. | Modeling and simulation of multi-scale environmental systems with generalized hybrid Petri nets | |
| CN116860221A (en) | Intelligent recommendation method and system for low-code data model | |
| CN102346815B (en) | Digital biological system for simulating biological competition and evolution process | |
| Schwefel | Deep insight from simple models of evolution | |
| Mainzer | Causality in natural, technical, and social systems | |
| Getz et al. | Using nova to construct agent-based models for epidemiological teaching and research | |
| Gray et al. | Increasing model efficiency by dynamically changing model representations | |
| Rouly | A computer simulation to investigate the association between gene-based gifting and pair-bonding in early hominins | |
| Baptista et al. | Step evolution: improving the performance of open-ended evolution simulations | |
| Bryndin | Formation and creative manifestation of functional ensembles of intellectual agents based on live information in various spheres of life activity | |
| Rosset et al. | Use of high performance computing in agent-based social simulation: A case study on trust-based coalition formation | |
| Kugler | Biocharts: Unifying biological hypotheses with models and experiments | |
| Xie et al. | Dynamical Analysis of a Commensalism System Incorporating Beddington-DeAngelis Functional Response and Allee Effect. | |
| Sloman | Design spaces, niche spaces and the “hard” problem | |
| Anderson | Creation of desirable complexity: strategies for designing selforganized systems |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20140820 Termination date: 20150926 |
|
| EXPY | Termination of patent right or utility model |