1.5
1.5

1

1
Output
y
out
0.5
0
–0.5
Output
y
out
0.5
0
–0.5
–1
–1
–1.5
–1.5
200
230
200
210
Time (s)
Time (s)
CPG1
CPG2
CPG3
CPG4
CPG5
CPG6
(a)
(b)
1.5
1.5
1
1
Output
y
out
0.5
0
–0.5
Output
y
out
0.5
0
–0.5
–1
–1
–1.5
–1.5
200
230
200
210
Time (s)
Time (s)
CPG1
CPG2
CPG3
CPG4
CPG5
CPG6
(a)
(b)
FIGURE 2.3
Rhythmic outputs of two kinds of CPG models using a unilaterally connected network: (a)
mutual inhibitory model and (b) cyclic inhibitory model.
19
CPG-Based Control of Serpentine Locomotion of a Snake-Like Robot
TABLE 2.1
Configuration of Parameters in the CPG Neuron
Driving input
u 0
2.5
Time constant
τ 1
2.0
Time constant
τ 2
6.0
Adaptation coefficient
β
2.5
Connection weight inner neurons
w
2.7
Connection weight among CPGs
w 0
0.1
FIGURE 2.4
Rhythmic outputs of two kinds of CPG models in a closed-loop network: (a) mutual inhibitory
model and (b) cyclic inhibitory model.
homogeneously distributed in one period (see Figure 2.4). Because the output of the CPG modules on the outside of the loop has the same phase as the
CPGs in the loop, the phase difference between the two neighboring CPG
modules in the whole network can be obtained by
Φ = 2π /U
(2.5)