A computational model for ratbot locomotion based on cyborg intelligence

作者:Zheng, Nenggan; Su, Lijuan; Zhang, Daqiang; Gao, Liqiang; Yao, Min*; Wu, Zhaohui
来源:Neurocomputing, 2015, 170: 92-97.
DOI:10.1016/j.neucom.2014.12.115

摘要

Ratbots with electric stimulation in their brains possess not only their own biological sensation, perception, memory, and locomotion control abilities, but also machine visual sensation, memory and computing functionalities. With electrodes implanted in the medial forebrain bundle (MFB), we propose here a hybrid bio-machine locomotion system in the ratbots, processing the machine visual inputs, forming hybrid multiple memory system and outputting locomotion commands for navigation behaviors. To illustrate the enhanced performance of the ratbots theoretically, a computational model is presented to show how the multiple memory system affects the central pattern generator (CPG) generating the gait pattern and running velocity. Compared with the extensive data from behavioral experiments, the results output from the proposed computational model fit the data of the decision accuracy and the relative velocity well, thus shown that the model provides a possible locomotion control mechanism innately in the cyborg systems.