Stochastic training of a biologically plausible spino-neuromuscular system model
Stanley Phillips Gotshall, Terence Soule · 2007
A primary goal of evolutionary robotics is to create systems that are as robust and adaptive as the human body. Moving toward this goal often involves training control systems that processes sensory information in a way similar to humans. Artificial neural networks have been an increasingly popular option for thisbecause they consist of processing units that approximate thesynaptic activity of biological signal processing units, i.e. neurons. In this paper we train a nonlinear recurrent spino-neuromuscular system model(SNMS) comparing the performance of genetic algorithms (GA)s, particle swarm optimizers (PSO)s, and GA/PSO hybrids. This model includes several key features of the SNMS that have previously been modeled individually but have not been combined into a single model as is done here. The results show that each algorithm produces fit solutions and generates fundamental biological behaviors that are not directly trained for such as tonic tension behaviors and tricepsactivation patterns.