Quadrupedal Robot Locomotion: A Biologically Inspired Approach and Its Hardware Implementation

Abstract

A bioinspired locomotion system for a quadruped robot is presented. Locomotion is achieved by a spiking neural network (SNN) that acts as a Central Pattern Generator (CPG) producing different locomotion patterns represented by their raster plots. To generate these patterns, the SNN is configured with specific parameters (synaptic weights and topologies), which were estimated by a metaheuristic method based on Christiansen Grammar Evolution (CGE). The system has been implemented and validated on two robot platforms; firstly, we tested our system on a quadruped robot and, secondly, on a hexapod one. In this last one, we simulated the case where two legs of the hexapod were amputated and its locomotion mechanism has been changed. For the quadruped robot, the control is performed by the spiking neural network implemented on an Arduino board with 35% of resource usage. In the hexapod robot, we used Spartan 6 FPGA board with only 3% of resource usage. Numerical results show the effectiveness of the proposed system in both cases.

Figure 1

Schematic drawing of different stereotypic quadrupedal walking patterns. In trot, two diagonal legs swing in synchrony (a). In walk, synchronous swing of a diagonal pair of legs is followed by two single leg swing phases (b, c). Black bars indicate leg swing and R and L correspond to right and left sides, respectively [8].

Figure 2

Modified locomotion patterns for the legged robots.

Figure 3

Schematic diagram of the CGE-based methodology for designing CPGs.

Figure 4

Hardware used in this research: (a) BotBoarduino, (b) FPGA, (c) servo controller SSC32, and (d) breakout board BRK6110.

Figure 5

Neuronal configurations in the legged robots. (a) Quadruped and (b) hexapod robots. C and F indicate coxa and femur, respectively. L and R correspond to the side where the neurons are located in the robot (images from Lynxmotion website).

Figure 6

(a) Network topology and (b) raster plot for the walking gait.

Figure 7

(a) Network topology and (b) raster plot for the jogging gait.

Figure 8

(a) Network topology and (b) raster plot for the running gait.

Figure 9

Real time simulation on a quadruped robot.

Figure 10

Real time simulation on a hexapod robot with middle legs amputated.

Figure 11

Network topologies with restrictions on the number of synaptic connections for (a) walking, (b) jogging, and (c) running gaits.

Figure 12

(a) All-in-one topology network. (b) The three locomotion gaits are generated by the same network.

Figure 13

Oscilloscope screenshots for the three locomotion gaits (real time monitoring). On the left side, we show the real time simulation for the quadruped robot, and on the right side the real time simulation for the amputated hexapod robot.

Algorithm 1

CPG design methodology.