List of Figures
1.1 Bottom-up and top-down approaches. . . . . . . . . . . . . . 2
1.2 Turing test. . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3 A Chines e room. . . . . . . . . . . . . . . . . . . . . . . . . 5
1.4 Simulation of humanoid robot motions. . . . . . . . . . . . . 7
1.5 Real-world motions of a humanoid robot. . . . . . . . . . . . 7
1.6 Mesa Verde Natio nal Park. . . . . . . . . . . . . . . . . . . . 9
2.1 GTYPE and PTYPE. . . . . . . . . . . . . . . . . . . . . . 15
2.2 GA op e rators. . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.3 GA cros sover s: (a) one-point crossove r, (b) n-point crossover,
and (c) uniform crossover. . . . . . . . . . . . . . . . . . . . 18
2.4 Examples of good and bad initialization. . . . . . . . . . . . 20
2.5 TSP simulator. . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.6 TSP crossover example. . . . . . . . . . . . . . . . . . . . . 23
2.7 Results of TSP simulation. . . . . . . . . . . . . . . . . . . . 24
2.8 Example of a (a) function and (b) program. . . . . . . . . . 26
2.9 Example of a tree structure. . . . . . . . . . . . . . . . . . . 26
2.10 Initialization of a tree structure using FULL and GROW
methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
2.11 Initial population obtained with the RAMPED HALF &
HALF method. . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.12 An example o f the wall following problem. . . . . . . . . . . 31
2.13 Genetic operations in GP. . . . . . . . . . . . . . . . . . . . 32
2.14 Robot s e nsors. . . . . . . . . . . . . . . . . . . . . . . . . . . 34
2.15 Wall following by GP. . . . . . . . . . . . . . . . . . . . . . 35
2.16 IEC algorithm. . . . . . . . . . . . . . . . . . . . . . . . . . 36
2.17 CG synthesis of plants based on L-systems. . . . . . . . . . 37
2.18 An overview of LGPC for art. . . . . . . . . . . . . . . . . . 39
2.19 An example r ock rhythm generated by CONGA. . . . . . . 40
2.20 A four -handed performance on the piano and Swarm. . . . . 43
3.1 Bottom-up model. . . . . . . . . . . . . . . . . . . . . . . . . 46
3.2 Swarm modeling. . . . . . . . . . . . . . . . . . . . . . . . . 46
3.3 Object-oriented simulation. . . . . . . . . . . . . . . . . . . 47
3.4 Model and observer. . . . . . . . . . . . . . . . . . . . . . . 48
3.5 Swarm as a virtual c omputer. . . . . . . . . . . . . . . . . . 49
xi
xii List of Figures
3.6 Elements of Activity library. . . . . . . . . . . . . . . . . . . 49
3.7 Recursive structures in Swarm. . . . . . . . . . . . . . . . . 50
3.8 Probes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
3.9 Probe and GUI. . . . . . . . . . . . . . . . . . . . . . . . . . 51
3.10 GUI widgets. . . . . . . . . . . . . . . . . . . . . . . . . . . 52
3.11 Class hierarchy in simpleSwarmBug. . . . . . . . . . . . . . 54
3.12 Execution seq uence in simpleSwarmBug. . . . . . . . . . . . 57
3.13 simpleSwarmBug2. . . . . . . . . . . . . . . . . . . . . . . . 58
3.14 Display of object distribution (ZoomRaster). . . . . . . . . . 61
3.15 Display with Swarm. . . . . . . . . . . . . . . . . . . . . . . 61
3.16 simpleObserverBug. . . . . . . . . . . . . . . . . . . . . . . . 62
3.17 Swarm hierarchy. . . . . . . . . . . . . . . . . . . . . . . . . 64
3.18 Class hierarchy in simpleObserverBug2. . . . . . . . . . . . . 65
3.19 Probes in simpleObser verBug2. . . . . . . . . . . . . . . . . 65
3.20 Objects and probes. . . . . . . . . . . . . . . . . . . . . . . . 66
3.21 simpleObserverBug2. . . . . . . . . . . . . . . . . . . . . . . 67
3.22 Probes for Bug. . . . . . . . . . . . . . . . . . . . . . . . . . 68
3.23 Execution seq uence in simpleExperBug. . . . . . . . . . . . 69
3.24 ExperSwarm. . . . . . . . . . . . . . . . . . . . . . . . . . . 71
4.1 Sexual se lec tion. . . . . . . . . . . . . . . . . . . . . . . . . . 75
4.2 The handicap principle. . . . . . . . . . . . . . . . . . . . . 76
4.3 The parasite principle. . . . . . . . . . . . . . . . . . . . . . 77
4.4 Kirkpatrick model and its equilibrium state. . . . . . . . . . 79
4.5 Evolutionary results after 500 generations. . . . . . . . . . . 80
4.6 Evolutionary transition in the case where a new gene is in-
serted. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
4.7 Survival ratios (male) v s. generatio ns. . . . . . . . . . . . . 82
4.8 Swarm simulation based on Kirkpatrick’s model. . . . . . . 84
4.9 Sexual se lec tion in a two-dimensional space. . . . . . . . . . 85
4.10 The prisoner’s dilemma. . . . . . . . . . . . . . . . . . . . . 86
4.11 IPD s imulation. . . . . . . . . . . . . . . . . . . . . . . . . . 100
4.12 Evolutionary ka leidoscope. . . . . . . . . . . . . . . . . . . . 103
4.13 Designed examples of genotype graphs and corresponding
creature morphologies. . . . . . . . . . . . . . . . . . . . . . 105
4.14 Creatures evolved for swimming. . . . . . . . . . . . . . . . 106
4.15 Evolved compe ting creatures. . . . . . . . . . . . . . . . . . 106
4.16 Various individuals. . . . . . . . . . . . . . . . . . . . . . . . 107
4.17 Best individua ls with generations. . . . . . . . . . . . . . . . 107
4.18 The suboptimal individual and the movement. . . . . . . . . 108
4.19 Cross-shaped robot. . . . . . . . . . . . . . . . . . . . . . . . 108
4.20 Wheel robot. . . . . . . . . . . . . . . . . . . . . . . . . . . 108
4.21 The bes t evolved individuals. . . . . . . . . . . . . . . . . . 109
5.1 Ant trail. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
List of Figures xiii
5.2 Bridge-shaped paths. . . . . . . . . . . . . . . . . . . . . . . 113
5.3 The ratio of ants that used the shorter path. . . . . . . . . . 113
5.4 Pheromone trails of ants. . . . . . . . . . . . . . . . . . . . 115
5.5 Path selection rules of ants. . . . . . . . . . . . . . . . . . . 116
5.6 TSP simulator by ACO. . . . . . . . . . . . . . . . . . . . . 118
5.7 Ant-clus tering. . . . . . . . . . . . . . . . . . . . . . . . . . 1 20
5.8 Ant-clus tering with Swarm. . . . . . . . . . . . . . . . . . . 121
5.9 Routing table. . . . . . . . . . . . . . . . . . . . . . . . . . . 122
5.10 Agent-based routing. . . . . . . . . . . . . . . . . . . . . . . 122
5.11 Comparison of AntNet and o ther routing methods (from [15]). 124
5.12 A scene of building a living bridge by army ants. . . . . . . 126
5.13 Simulation environment. . . . . . . . . . . . . . . . . . . . . 12 7
5.14 Swarm-based simulation of army ants. . . . . . . . . . . . . 128
5.15 State tr ansition of a gents. . . . . . . . . . . . . . . . . . . . 129
5.16 Maps for experiment. . . . . . . . . . . . . . . . . . . . . . . 129
5.17 Simple map – experimental results. . . . . . . . . . . . . . . 131
5.18 Difficult ma p – experimental results 1. . . . . . . . . . . . . 131
5.19 Difficult ma p – experimental results 2. . . . . . . . . . . . . 132
5.20 Perfo rmance co mparison in terms of foraging time. . . . . . 133
5.21 Perfo rmance co mparison in terms of altruistic activity. . . . 133
5.22 Comparison of bridge co nstruction sites. . . . . . . . . . . . 1 34
5.23 Maps used to study the effect of the number of agents. . . . 135
5.24 Effect of neighborhood knowledge (Map 1). . . . . . . . . . 136
5.25 Effect of neighborhood knowledge (Map 2). . . . . . . . . . 136
5.26 Changes in size o f bridge. . . . . . . . . . . . . . . . . . . . 137
5.27 Changes in size o f chain. . . . . . . . . . . . . . . . . . . . . 137
5.28 Experimental results with task assignment. . . . . . . . . . . 138
6.1 Do the movements of a school of fish follow a certain set of
rules? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
6.2 Simple behavior of b oids ((a)⇒(b)⇒(c)⇒(d)). . . . . . . . 141
6.3 Boids in a situatio n with obstacles ((a)⇒(b)). . . . . . . . . 141
6.4 Avoid collision (1). . . . . . . . . . . . . . . . . . . . . . . . 1 42
6.5 Avoid collision (2). . . . . . . . . . . . . . . . . . . . . . . . 1 42
6.6 Ahead or behind a line that crosses the boid’s eyes. . . . . . 142
6.7 Updating the velocity vector. . . . . . . . . . . . . . . . . . 143
6.8 Each boid has its own field of view. . . . . . . . . . . . . . . 144
6.9 Behavior in Swarm Chemistry (1). . . . . . . . . . . . . . . 147
6.10 Behavior in Swarm Chemistry (2). . . . . . . . . . . . . . . 147
6.11 A snapshot of Swarm Chemistry. . . . . . . . . . . . . . . . 148
6.12 Flow chart of the PSO algorithm. . . . . . . . . . . . . . . . 150
6.13 In which way do birds fly? . . . . . . . . . . . . . . . . . . . 151
6.14 PSO simulator. . . . . . . . . . . . . . . . . . . . . . . . . . 152
6.15 Concept of searching process by PSO with a Gaussian muta-
tion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
xiv List of Figures
6.16 Rastrigin’s function (F 8). . . . . . . . . . . . . . . . . . . . 154
6.17 Griewangk’s function (F 9). . . . . . . . . . . . . . . . . . . 154
6.18 Standard PSO versus PSO with a Gaussian mutation for F 8. 155
6.19 Standard PSO versus PSO with a Gaussian mutation for F 9. 156
6.20 A bee colony. . . . . . . . . . . . . . . . . . . . . . . . . . . 157
6.21 Waggle dance. . . . . . . . . . . . . . . . . . . . . . . . . . . 158
6.22 ABC simulator with Swarm. . . . . . . . . . . . . . . . . . . 161
6.23 ABC optimization with Swarm. . . . . . . . . . . . . . . . . 162
6.24 Bug world: (a)166th generation, (b) 39, 618th generation. . . 165
6.25 Bug’s ge ne code. . . . . . . . . . . . . . . . . . . . . . . . . 166
6.26 A schematic illustration of bug s. . . . . . . . . . . . . . . . . 167
6.27 Ty pes of bugs. . . . . . . . . . . . . . . . . . . . . . . . . . . 168
6.28 Perfo rmance co mparison. . . . . . . . . . . . . . . . . . . . . 169
6.29 Garden of Eden (a) 69th generation (b) 72, 337th generation
(c) 478, 462nd generation. . . . . . . . . . . . . . . . . . . . 170
6.30 Bugs’ motions for F 2: (a) 5th generation, (b) 13th generation,
(c) 25th generation, (d) 53rd generation. . . . . . . . . . . . 174
6.31 Illustration o f bug-based se arch. . . . . . . . . . . . . . . . . 177
6.32 BUGS simulator. . . . . . . . . . . . . . . . . . . . . . . . . 178
7.1 CA c arrying out majority voting. . . . . . . . . . . . . . . . 182
7.2 Behavior of CA driven by GA. . . . . . . . . . . . . . . . . . 182
7.3 Explanation of CA behavior from collision of particles. . . . 184
7.4 Examples of pa tter ns. . . . . . . . . . . . . . . . . . . . . . 185
7.5 Congestion simulation using BCA. . . . . . . . . . . . . . . 187
7.6 Simulation of silicon traffic. . . . . . . . . . . . . . . . . . . 188
7.7 Game of life. . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
7.8 One-dimensional cellular a utomaton. . . . . . . . . . . . . . 191
7.9 Execution of Wolfram’s expe riment. . . . . . . . . . . . . . . 192
7.10 Self-replicating loop of Langton. . . . . . . . . . . . . . . . . 195
7.11 Forest fire e xamples. . . . . . . . . . . . . . . . . . . . . . . 199
7.12 Examples of forest fires. . . . . . . . . . . . . . . . . . . . . 201
7.13 Schelling’s simulation of the se gregation model. . . . . . . . 203
7.14 An example o f a co llision process in the HPP model
((a)→(b)→(c)→(d)). . . . . . . . . . . . . . . . . . . . . . . 205
7.15 Collision pr ocess in the FHP model. . . . . . . . . . . . . . . 206
7.16 Simulation examples of rain drops using the LGA method. 207
7.17 LGA simulation with the HPP model. . . . . . . . . . . . . 208
7.18 An example o f s imulation using the LGA method. . . . . . 208
7.19 CA patter ns found on shells. . . . . . . . . . . . . . . . . . 209
7.20 Turing model simulation results. . . . . . . . . . . . . . . . . 211
7.21 Parameters probe for the Turing model. . . . . . . . . . . . 21 2
7.22 The whole structure for percolation. . . . . . . . . . . . . . 214
7.23 patternSpace: the da rk portion is the largest cluster. . . . . 214
7.24 Percolation probability. . . . . . . . . . . . . . . . . . . . . . 214
List of Figures xv
7.25 Histograms in Swarm. . . . . . . . . . . . . . . . . . . . . . 215
7.26 An example o f the flow of cars. . . . . . . . . . . . . . . . . 216
7.27 Simulating a traffic jam (with SlowtoStart). . . . . . . . . . 21 7
7.28 Simulating a traffic jam (without SlowtoStart). . . . . . . . 218
7.29 ASEP model s imulation. . . . . . . . . . . . . . . . . . . . . 218
7.30 Two sugar (green) mountains in the Sugarscape model. . . . 219
7.31 Agent aggregation in (G
1
, M). . . . . . . . . . . . . . . . . 2 21
7.32 Popula tion and feature changes in (G
1
, M). . . . . . . . . . 221
7.33 Features (vision, metabolism) in (G
1
, {M, R
[60,100]
}) . . . . 223
7.34 Wealth in (G
1
, {M, R
[60,100]
}). . . . . . . . . . . . . . . . . 223
7.35 Features in (G
1
, {M, R
[60,100]
, S}). . . . . . . . . . . . . . . 225
7.36 Wealth in (G
1
, {M, R
[60,100]
, S}). . . . . . . . . . . . . . . . 225
7.37 Agent aggregation due to seasonal variation:
(S
50,8,1
, {M, R
[60,100]
, S}). . . . . . . . . . . . . . . . . . . . 227
7.38 Popula tion changes due to seasonal variation:
(S
50,8,1
, {M, R
[60,100]
, S}). . . . . . . . . . . . . . . . . . . . 227
7.39 Feature and wealth changes due to seasonal variation:
(S
50,8,1
, {M, R
[60,100]
, S}). . . . . . . . . . . . . . . . . . . . 228
7.40 Agent aggregation due to pollution:
({G
1
, D
1
}, {M, P
1,1
}). . . . . . . . . . . . . . . . . . . . . . 229
7.41 Feature and wealth changes due to pollution:
(S
50,8,1
, {M, R
[60,100]
, S}). . . . . . . . . . . . . . . . . . . . 230
7.42 Agent aggregation with culture propa gation (100 steps later):
(G
1
, {M, K}). . . . . . . . . . . . . . . . . . . . . . . . . . . 231
7.43 Diffusion process of two tr ibes : without culture propagation. 232
7.44 Diffusion pr ocess of two tribes: with culture propagation (100
steps later). . . . . . . . . . . . . . . . . . . . . . . . . . . . 232
7.45 Diffusion process of two tribes: with culture propagation (no.
of tags = 7; 200 steps later). . . . . . . . . . . . . . . . . . . 233
7.46 Introduction of combat (without culture propagation):
(G
1
, C
∞
). . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235
7.47 Introduction of combat (with culture propagation): (G
1
, C
∞
). 236
7.48 Combat intro duced at the 200th step. . . . . . . . . . . . . . 23 6
7.49 Introduction of spice. . . . . . . . . . . . . . . . . . . . . . . 237
7.50 Agent aggregation with spice: (G
1
, M). . . . . . . . . . . . . 2 39
7.51 MRS values with spice: (G
1
, M). . . . . . . . . . . . . . . . 239
7.52 Agent aggregation with spice and combat. . . . . . . . . . . 240
7.53 Agent aggregation with trade. . . . . . . . . . . . . . . . . . 242
7.54 Traded amounts and prices. . . . . . . . . . . . . . . . . . . 242
7.55 Trading prices and reserves in the case of seasonal change and
trade. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
7.56 Agent aggregation when only red agents are capable of trad-
ing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
7.57 Sugarscape in Swarm. . . . . . . . . . . . . . . . . . . . . . 245
..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset
3.146.176.68