2.6 A Spatio-Temporal Algorithm for Controlling Locomotion of a Hexapod Robot

We focus now on a specific example, the design of a CPG for a hexapod robot that accounts for the several different gaits that hexapods adopt in different environmental conditions. We discuss an algorithm accounting for the generation of distinct gaits and show simulation results.

Hexapod walkers typically adopt several gaits characterized by different speeds: at low speeds the so-called slow gait or wave gait is adopted, medium gait emerges at intermediate speeds, and alternating tripod or fast gait is adopted at high speeds [Collins and Stewart (1993b); Pearson (1976); Wilson (1966)].

Using the parameters previously defined of duty cycle and phase lags, it is possible to characterize fast gait as a periodic gait with the same duty factor for all the legs ( df _i=0.5) and the following phase lags:

(2.19)

For medium gait the duty factor is df _i=5/8 and the phase lags are as follows:

(2.20)

For slow gait the duty factor is df _i=9/12 and the phase lags are as follows:

(2.21)

These three gaits are obtained by choosing the connections among the six neurons as in Fig. 2.10. Since the symbols adopted are not standard, it should be pointed out that each circle represents a CPG cell as in Eq. (2.5) controlling a hexapod leg. Moreover, the connections indicate the synapses realizing the CPG scheme: inhibitory synapses (i.e. with a negative sign) terminate with a dot,...