智能算法与应用探索：A*寻路与接纳采样

需积分: 9 30 浏览量更新于2024-07-22 收藏 1.14MB PDF 举报

"智能加工和应用" 本文档主要探讨的是智能技术在模块设计和应用中的原理，尤其关注A*算法和接受-拒绝方法等优化策略。A*算法是人工智能领域中的一种启发式搜索方法，它在搜索树中选择一个节点进行扩展，这个节点的代价是到达该节点的成本加上该节点的启发式成本值，而启发式成本是对完成任务的真实最小成本的低估。这种方法在路径规划、游戏AI以及许多其他问题求解中有着广泛应用。 Hart等人在1968年的论文中为启发式确定最低成本路径提供了一个正式的基础，这篇论文是理解A*算法的重要参考资料。而Pearl在1984年的《Heuristics》一书中进一步深入讨论了启发式方法的理论与实践。此外，文档还提到了接受-拒绝方法，这是在随机或蒙特卡洛模拟中用于从难以直接采样的目标概率分布中抽取样本的一种策略。首先，从一个与目标分布相近且易于采样的分布中生成样本，然后根据一定的条件可能拒绝这些样本。这种方法在统计模拟、经济模型构建以及复杂系统分析等领域有广泛的应用。接着，文档提到了会计价格和影子价格，会计价格通常指传统财务会计中的资产价值，而影子价格则是在优化问题中，对资源限制的非市场价值的估计，常用于决策支持和资源配置。模型认证（Accreditation）可能是指对模型的性能、准确性和可靠性进行评估的过程，这对于确保智能系统在实际应用中的有效性至关重要。 S.I. 可能指的是系统识别(System Identification)，这是一个领域，研究如何从数据中识别和建立系统的数学模型，这在控制理论、机器学习和数据分析中都是关键步骤。这份智能英文文献涵盖了人工智能中的核心算法、模拟策略以及经济与系统分析的关键概念，对于理解和应用智能技术具有很高的参考价值。深入研究这些内容将有助于提升在智能系统设计和优化方面的专业知识。

Bedau, M. A., & Humphreys, P. (Eds.). (2007). Emergence:

Contemporary readings in philosophy and science. London:

MIT Press.

Bonabeau, E. (2001). Agent-based modeling: Methods and

techniques for simulating human systems. Proceedings of

the National Academy of Sciences of the United States of

America, 99(3), 7280–7287.

Bonabeau, E., Dorigo, M., & Theraulaz, G. (1999). Swarm

intelligence: From natural to artiﬁcial systems. New York:

Oxford University Press.

Brown, D., et al. (2005). Spatial process and data models:

Toward integration of agent-based models and GIS.

Journal of Geographical Systems, 7(1), 25–47.

Carley, K., et al. (2006). Biowar: Scalable agent-based model of

bioattacks. IEEE Transactions on Systems, Man, and

Cybernetics - Part A: Systems and Humans, 36(2), 252–265.

Epstein, J. (2009). Modelling to contain pandemics. Nature,

460(6), 687.

Epstein, J., & Axtell, R. (1996). Growing artiﬁcial societies:

Social science from the bottom up. Cambridge, MA: MIT

Press.

Germann, T., Kadau, K., Longini, I., & Macken, C. (2006).

Mitigation strategies for pandemic inﬂuenza in the United

States. Proceedings of the National Academy of Sciences,

103(15), 5935–5940.

Gilbert, N., & Troitzsch, K. G. (1999). Simulation for the social

scientist. Buckingham, UK: Open University Press.

Grimm, V., et al. (2006). A standard protocol for describing

individual-based and agent-based models. Ecological

Modelling, 198(1–2), 115–126.

Holland, J. H. (1975). Adaptation in natural and artiﬁcial

Systems. Ann Arbor, MI: University of Michigan Press.

Holland, J., & Miller, J. H. (1991). Artiﬁcial adaptive agents in

economic theory. The American Economic Review, 81(2),

365–371.

Kohler, T. A., Gumerman, G. J., & Reynolds, R. G. (2005).

Simulating ancient societies. Scientiﬁc American.

Langton, C. G. (1989). Artiﬁcial life. In C. G. Langton (Ed.),

Artiﬁcial life: The proceedings of an interdisciplinary

workshop on the synthesis and simulation of living systems

(pp. 1–47). Reading, MA: Addison-Wesley.

Macal, C. M., & North, M. J. (2011). Introductory tutorial:

Agent-based modeling and simulation. In S. Jain, R. R.

Creasey, J. Himmelspach, K. P. White, & M. Fu (Eds.),

Proceedings of the 2011 Winter Simulation Conference

(pp. 1456–1469).

Macal, C. M., & North, M. J. (2010). Tutorial on agent-based

modelling and simulation. Journal of Simulation, 4(3),

151–162.

Macy, M. W., & Willer, R. (2002). From factors to actors:

Computational sociology and agent-based modeling.

Annual Review of Sociology, 28, 143–166.

Manson, S. M. (2006). Bounded rationality in agent-based

models: Experiments with evolutionary programs.

International Journal of Geographical Information Science,

20(9), 991–1012.

National Research Council. (2008). Behavioral modeling and

simulation: From individuals to societies . Washington, DC:

National Academies Press.

Niazi, M., Hussain, A. & Kolberg, M. (2009). Veriﬁcation and

validation of agent-based simulations using the VOMAS

approach. Proceedings of the Third Workshop on Multi-

Agent Systems and Simulation ’09 (MASS ’09). Sep 7–11,

2009. Torino, Italy.

North, M. J., & Macal, C. M. (2007). Managing business

complexity: Discovering strategic solutions with agent-

based modeling and simulation. Oxford, UK: Oxford

University Press.

Pritsker, A. A. B. (1979). Compilation of deﬁnitions of

simulation. Simulation, 33(2), 61–63.

Rand, W., & Rust, R. T. (2011). Agent-based modeling in

marketing: Guidelines for rigor. International Journal of

Research in Marketing, 28(3), 181–193.

Sakoda, J. M. (1971). The checkerboard model of social

interaction. Journal of Mathematical Sociology, 1, 119–132.

Samuelson, D. (2000). Designing organizations. OR/MS Today,

27(6).

Samuelson, D., & Macal, C. (2006). Agent-based mode ling

comes of age. OR/MS Today, 33(4), 34–38.

Samuelson, D., et al. (2007). Agent-based simulation of mass

egress after an improvised explosive device attack.

Homeland Security Institute Final Report to the Department

of Homeland Security, Science and Technology Directorate.

HSI Document Number RP06-IOA-31-03.

Samuelson, D., et al. (2010). Agent-based simulations of mass

egress after an IED attack. In W. Klingsch, C. Rogsch,

A. Schadschneider, & M. Schreckenberg (Eds.), Pedestrian

and evacuation dynamics 2008 (PED2008). London/

New York: Springer.

Schelling, T. C. (1971). Dynamic models of segregation. Journal

of Mathematical Sociology, 1, 143–186.

Simon, H. A. (1997). Behavioral economics and bounded

rationality. In H. A. Simon (Ed.), Models of bounded

rationality (pp. 267–298). Cambridge, MA: MIT Press.

Sun, R. (2006). Cognition and multi-agent interaction: From

cognitive modeling to social simulation. Cambridge, UK:

Cambridge University Press.

Tesfatsion, L., & Judd, K. L. (Eds.). (2006). Handbook of

computational economics, volume II: Agent-based

computational economics. Amsterdam: Elsevier/North-Holland.

Ziegler, B. P., Praehofer, H., & Kim, T. G. (2000). Theory of

modeling and simulation (2nd ed.). San Diego, CA:

Academic Press.

Agriculture and the Food Industry

Filmore Bender

and Gerald Kahan

University of Maryland, College Park, MD, USA

McCormick and Company, Sparks, MA, USA

It is often difﬁcult to determine where the agricultural

sector of an economy ends and the nonagricultural

sector begins. For the purpose of this article, the

agricultural sector of the economy is deﬁned as

production and supply of agricultural inputs, the

A 16 Agriculture and the Food Industry

production of agricultural goods on farms and ranches,

the processing and transportation of those goods, as

well as the wholesaling and retailing of ﬁnished

products. Deﬁned in this way, the agricultural sector

of the economy in the United States represents

approximately 24% of the gross national product.

As with the nonagricultural sector of the economy,

operations research was ﬁrst used to solve agricultural

problems in the 1940s and 1950s. A Survey of

Agricultural Economics Literature, Vol. 2:

Quantitative Methods in Agricultural Economics,

1940s to 1970s traces the development of operations

research in addressing problems of importance to

agriculture (Judge et al. 1977). This work ranged

from quantifying production functions to the

development of models, simulation structures and the

use of linear programming and nonlinear optimization

models to quantify or predict economic consequ ences,

or alternatively, the use of these tools to solve speciﬁc

problems for speciﬁc ﬁrms.

Different components of the agricultural economy

have embraced the tools of operations research with

different levels of enthusiasm. Those sectors of

agriculture that can exercise considerable control of

inputs and environmental factors (e.g., feeder cattle,

broilers, eggs, pork and dairy production) began

adopting the tools of operations research during the

late 1950s. By 1965, essentially all of the feed

formulated for poultry in the United Sates was done

using least-cost linear-programming feed

formulations. Simultaneously, during the 1960s, the

beef industry began to adopt linear programmi ng on

a limited basis for least-cost feed formulation and for

the development of optimal production and marketing

strategies. The use of linear programming for least-

cost dairy rations became standard practice during the

late 1970s. Forestry, which is a branch of agriculture,

uses multi-period linear programming models to

determine optimal planting and harvesting schedules.

A number of interesting examples have been

reported in the literature which describe how linear

programming was used to solv e a variety of

agriculture problems. Upc raft et al. (1989) reported

that the soil water deﬁcit is the main decision

variable that British farmers monitor to decide when

to irrigate a particular ﬁeld and how much water to use.

The decision is generally based on the soil water deﬁcit

in the ﬁrst strip to be irrigated within each ﬁeld.

A mixed-integer linear program was constructed to

model the short-term irrigation scheduling problem

for a hose-reel/rain-gun irrigation system. Optimal

schedules were produc ed by quantifying the costs and

beneﬁts of irrigation, subject to the constraints of

equipment, labor, and availability of water. The

model is uniqu e in producing whole farm-irrigation

schedules, rather than individual ﬁeld schedules for

hose-reel/rain-gun irrigation systems.

The efﬁcient operation of a beef cattle feedlot is

controlled by the pri ce of the animals, purchase and

selling weights, and the feeding system . The optimal

feeding system involves feeding least-cos t rations to

animals at each stage in the production process. Glen

(1980) reported the development of an optimal method

for determining optimal feeding systems that meet the

nutrient standards recommended by the US National

Research Council. The approach involved using linear

programming to determine the least-cost rations to

produce speciﬁed live weight gains in animals of

known live weight. Dynamic programming was used

to determine the optimal sequence of rations to feed to

produce animals of speciﬁed live weight from known

live weight at minimum cost, using least-cost rations

from the linear programming model. Results from the

dynamic programming mode l can be used to determine

the optimal combination of purchase weight, selling

weight, and feeding system. The linear programming

model must be solved a large number of times to use

the dynam ic programming model.

In assessing feeding policy in livestock production,

it is generally assumed that an optimal feeding policy

will involve using least-cost rations throughout the

production process. Glen (1980) showed that this

assumption may not always be valid, particularly

when the supply of some of the feedstuffs used for

feeding the livestock is limited. A technique for

testing the validity of this assumption was presented

using a linear programming model of an integrated

crop and intensive beef production enterprise in

which some of the crops are used for livestock

feeding. An interactive solution procedure was

proposed for cases where this assumption was not

valid. While the computational burden associated

with the procedure for ﬁnding an improved solution

is large, experience with realistic data suggests that the

results from the linear programming model are likely

to be optimal.

Intra-year milk supply patterns depend largely on the

distribution of cow calving dates which are in turn

Agriculture and the Food Industry 17 A

inﬂuenced by climatic conditions. The most important

and least-costly input to milk production is the fresh

growth and high digestibility of grass in spring and

early summer that often gives rise to a highly seasonal

distribution of calving resulting in a seasonal milk supply

pattern. However, milk for liquid consumption and

production of perishable milk products must be geared

to meet a constant consumer demand throughout the

year, which necessitates a considerable amount of

production outside the least-cost period. Killen and

Keane (1978) re-ported on the development of a linear

programming model that gives the distribution of calving

dates, minimizes production costs, and meets consumer

demand for milk and related products. In addition, the

dual solution gives a set of seasonal prices which should

be paid to producers, equitably compensating them for

the costs they incur.

The agricultural sector deals with a biological

system. By its very nature, agriculture has elements

that foster the use of operations research techniques

and other elements that greatly impede the application

of these tools. Because many agricultural produc tion

units are relatively small in size, they are unable to

adopt operations research techniques in a cost effective

manner. On the other hand, because agricultural ﬁrms

are dispersed, those ﬁrms that either supply inputs to

farms or harvest and process agricultural products can

make effective use of truck routing and other spatial

optimization techniques.

Because of the savings in costs that can be achieved,

as well as the increasing availability of computers and

software, it is reasonable to expect increasing use of

the tools of operations research in agriculture. In fact,

as early as 1973, Beneke and Winterboer published

Linear Programming Applications to Agriculture,

a book devoted exclusively to the use of linear

programming in agriculture.

In the food industry, linear programming is

becoming increasingly common. Publications have

re-ported the use of linear programming in

formulating preblended meats (Rust 1976); luncheon

or sandwich meat (IBM 1966 ; Wieske 1981);

a protein-enriched luncheon sausage (Nicklin 1979);

bologna (IBM 1966); frankfurters (IBM 1966); and

a variety of sausage products (MacKenzie 1964; IBM

1966; Skinner et al. 1969). Ice cream is another food

product which has been successfully formulated

using linear programming (IBM 1964; Dano 1974;

Singh et al. 1979).

Cereal-based food blends have been formulated

using linear programming to insure adequate levels of

good-quality protein. Since these blends are sometimes

shipped to developing countries, linear programming

has helped to ensure that the prominent grain of the

country is present in the blend as a major ingredient. It

is desirable to blend cereal grains, since plant proteins

are usually deﬁcient in one or more of the essential

amino acids. Inglett et al. (1969)usedlinear

programming to bring the essential-amino-acid pattern

of a cereal-based food as close as possible to the pattern

found in a hen’s egg. Cavins et al. (1972)usedlinear

programming to formulate a least-cost cereal-based

food. The protein quality was controlled by setting

both lower and upper limits on each essential amino

acid in terms of its percent of total essential amino acid

content. Hsu et al. (1977a, b) studied the blending of

a wide range of plant and animal protein sources in

formulations for bread, pasta, cookies, and extruded

cornmeal snack and sausage. Constraints were used to

restrict both the nutritional and functional properties.

Roush et al. (1994

) reported on using chanc e

constrained programming to formulate commercial

feeds for animals. Nutritiona l consistency of ﬁnished

feeds increased by 40 % while costs dropped com pared

to feeds formulated by linear programming with

a margin of safety. With the exception of the

probabilistic constraints, the objective function and

most other constraints were linear in this model.

A detailed description of the formulation of a

low-cholesterol, low-fat beef stew using linear

programming was given by Bender et al. (1976).

The objective was to minimize cost, while enforcing

nutritional and other constraints based on the

recommendations for fat-modiﬁed and low-cholesterol

diets. These constraints were for a 100-g portion of stew

and set an upper limit on cholesterol content; a lower

limit on protein, vitamin A, thiamin, riboﬂavin, niacin,

vitamin C, and iron; and both an upper and a lower limit

on carbohydrate, fat, and calories.

Dano (1974) provided an full description of the

application of linear programming to a beer-blending

problem, and Wieske (1981) described the formulation

of an optimal margarine product. Another

application has been the formulation of mayonnaise

(Bender et al. 1982).

An interesting problem in the production of

champagne was reported by Hruby and Panton

(1993). In one of two methods used to produce

A 18 Agriculture and the Food Industry

champagne in Australia, a base wine called tirage is

allowed to ferment and mature in a bottle for as long as

6 months. The tirage is then transferred to a tank for

2 weeks where it is further processed. Finally, fresh

bottles are ﬁlled with ﬁnished product and stored for

6 weeks. Uncertainties in consumer demand and

constraints on production stretched a 9 month process

into 12 months. In addition, inventories of maturing

and ﬁnished product were far too high. The problem

was solved when a time-staged linear-programming

model was used to smooth production and reduce

stock levels.

The feasibility of planning menus by computer was

generally established in the early 1960s (Balintfy and

Blackburn 1964), as was the feasibility of

computerized menu analysis (Brisbane 1964). In

these models, nutritional requirements were provided

at lowest cost. Developing models which meets

sensory objectives as well as nutritional requirements

has proved to be a much more difﬁcult problem.

Renaud and Yacout (1996) reported on a company

that processes lobster primarily for foreign markets.

With increasing international competition, stricter

standards and decreasi ng annual volume of lobster

catches, the company wanted to know the optimal

product mix to maximize proﬁt. Linear programming

was used to generate ﬁve scenarios that encompassed

different possibilities available to management.

See

▶ Linear Programming

▶ Natural Resources

▶ Stigler’s Diet Problem

▶ Vehicle Routing

References

Balintfy, J. L., & Blackburn, C. R. (1964). From New Orleans:

A signiﬁcant advance in hospital menu planning by

computer. Institutions Magazine, 55(1), 54.

Bender, F. E., Kramer, A., & Kahan, G. (1976). Systems analysis

for the food industry. Westport, Connecticut: AVI

Publication.

Bender, F. E., Kramer, A., & Kahan, G. (1982). Linear

programming and its application in the food industry. Food

Technology, 36(7), 94.

Beneke, R. R., & Winterboer, R. D. (1973). Linear programming

applications to agriculture. Ames: Iowa State Press.

Brisbane, H. M. (1964). Computing menu nutrients by data

processing. Journal of the American Dietetic Association,

44, 453.

Cavins, J. F., Inglett, G. E., & Wall, J. S. (1972). Linear

programming controls amino acid balance in food

formulation. Food Technology, 26(6), 46.

Dano, S. (1974). Linear programming in industry (4th ed.).

New York: Springer.

Glen, J. J. (1980). A Mathematical programming approach to beef

feedlot optimization. Management Science, 26, 524–535.

Hruby, H. F., & Panton, D. M. (1993). Scheduling transfer

champagne production. International Journal of

Management Science, 21, 691–697.

Hsu, H. W., Satterlee, L. D., & Kendrick, J. G. (1977a).

Experimental design: Computer blending predetermines

properties of protein foods, part I. Food Product

Development, 11(7), 52.

Hsu, H. W., Satterlee, L. D., & Kendrick, J. G. (1977b).

Results and discussion: Computer blending predetermines

properties of protein foods, part II. Food Product

Development, 11(8), 70.

IBM. (1964). Linear programming — ice cream blending. White

Plains, NY: IBM Technical Publications Department.

IBM. (1966). Linear programming — meat blending. White

Plains, NY: IBM Technical Publications Department.

Inglett, G. E., Cavins, J. F., Kwokek, W. F., & Wall, J. S. (1969).

Using a computer to optimize cereal based food composition.

Cereal Science Today, 14(3), 69.

Judge, G. G., Day, R., JohnsonSR, R. G., & Martin, L. R. (1977).

A survey of agricultural economics literature (Quantitative

methods in agricultural economics, 1940s to 1970s, Vol. 2).

Minneapolis: University of Minnesota Press.

Killen, L., & Keane, M. (1978). A linear programming model of

seasonality in milk production. Journal of Operational

Research Society, 29, 625–631.

MacKenzie, D. S. (1964). Prepared meat product

manufacturing. Chicago: AMI Center for Continuing

Education, American Meat Institute.

Nicklin, S. H. (1979). The use of linear programming in food

product formulations. Food Technology in New Zealand,

14(6), 2.

Renaud, I., & Yacout, S. (1996). Resource allocation and

optimal product mix at a ﬁsh and seafood processing

company. Computers and Industrial Engineering,

31, 355–358.

Roush, W. B., Stock, R. H., Cravener, T. L., & D‘Alfonso, T. H.

(1994). Using chance-constrained programming for animal

feed formulation at Agway. Interfaces, 24(2), 53–58.

Rust, R. E. (1976). Sausage and processed meats manufacturing.

Washington, DC: AMI Center for Continuing Education.

American Meat Institute.

Singh, R. V., et al. (1979). Least cost ice-cream mix formulation:

A linear programming approach. Agricultural Situation in

India, 33(1), 7.

Skinner, R. H., et al. (1969). Food Industry applications of linear

programming. Food Manufacturing, 44(10), 35.

Upcraft, M. J., et al. (1989). A mixed linear programme for short-

term irrigation scheduling. Journal of Operational Research

Society, 40, 923–931.

Weske, R. (1981). Criteria of food acceptance (6th ed.). Zurich:

Forster-Verlag.

Agriculture and the Food Industry 19 A

剩余94页未读，继续阅读

baidu_28201105

粉丝: 0
资源: 2

智能算法与应用探索：A*寻路与接纳采样

人工智能在数控加工中的应用.pdf

人工智能加工中心机器人的研究.pdf

“华中数控杯”机器人装调与智能加工单元应用技能大赛在宁夏开幕.pdf

人工智能在矿物加工和选矿中的应用.pptx

人工智能在信息加工中的应用

智能加工技术在模具制造中的应用探索

关于智能加工系统的动态调度研究与应用.pdf

人工智能在石材加工中的应用.pptx

智能CNC加工自动编程系统的开发和应用可行性报告.pdf

智能CNC加工自动编程系统的开发和应用可行性报告.docx

最新资源