Modular Neural Control for Object Transportation of a Bio-inspired Hexapod Robot
Insects, like dung beetles, can perform versatile motor behaviors including walking, climbing an object (i.e., dung ball), as well as manipulating and transporting it. To achieve such complex behaviors for artificial legged systems, we present here modular neural control of a bio-inspired hexapod robot. The controller utilizes discrete-time neurodynamics and consists of seven modules based on three generic neural networks. One is a neural oscillator network serving as a central pattern generator (CPG) which generates basic rhythmic patterns. The other two networks are so-called velocity regulating and phase switching networks. They are used for regulating the rhythmic patterns and changing their phase. As a result, the modular neural control enables the hexapod robot to walk and climb a large cylinder object with a diameter of 18 cm (i.e., \(\approx 2.8\) times the robot’s body height). Additionally, it can also generate different hind leg movements for different object manipulation modes, like soft and hard pushing. Combining these pushing modes, the robot can quickly transport the object across an obstacle with a height up to 10 cm (i.e., \(\approx 1.5\) times the robot’s body height). The controller was developed and evaluated using a physical simulation environment.
KeywordsObject manipulation Locomotion Modular neural network Central pattern generator Walking machines Autonomous robots
We would like to thank Georg Martius for technical advise about the LpzRobots simulation software.
- 5.Rehman, B.U., Focchi, M., Frigerio, M., Goldsmith, J., Caldwell, D.G., Semini, C.: Design of a hydraulically actuated arm for a quadruped robot. In: Proceedings of the International Conference on Climbing and Walking Robots, pp. 283–290 (2015)Google Scholar
- 6.Heppner, G., Buettner, T., Roennau, A., Dillmann, R.: Versatile - high power gripper for a six legged walking robot. In: Proceedings of the International Conference on Climbing and Walking Robots, pp. 461–468 (2014)Google Scholar
- 8.Inoue, K., Ooe, K., Lee, S.: Pushing methods for working six-legged robots capable of locomotion and manipulation in three modes. In: Proceedings of the IEEE International Conference on Robotics and Automation, pp. 4742–4748 (2010)Google Scholar
- 9.Takeo, G., Takubo, T., Ohara, K., Mae, Y., Arai, T.: Internal force control for rolling operation of polygonal prism. In: Proceedings of the IEEE International Conference on Robotics and Biomimetics, pp. 586–591 (2009)Google Scholar
- 12.Valsalam, V., Miikkulainen, R.: Modular neuroevolution for multilegged locomotion. In: Proceedings of the Genetic and Evolutionary Computation Conference, pp. 265–272 (2008)Google Scholar
- 17.Pasemann, F., Hild, M., Zahedi, K.: So(2)-networks as neural oscillators. In: Proceedings of 7th International Work-Conference on Artificial and Natural Neural Networks (IWANN 2003), pp. 1042–1042 (2003)Google Scholar