Research On The Development Of Robot With A Bionic Learning Algorithm

In this paper, aiming at the problem of self-balancing control of two-wheeled robot and path planning in unknown environment, based on concept of intrinsic motivation in psychology as the internal driving force, proposed a bionic development algorithm based on intrinsic motivation. By using Intrinsic motivation based on psychological inspiration to replace the reward mechanism, this algorithm, which adopts different learning framework, can be used to improve the learning initiative in the unknown environment.The main research achievements of this thesis are as follows:1) Reinforcement learning algorithm based on intrinsic motivationOn the basis of the research of developmental psychology, neuroscience and developmental robotics, a reinforcement learning algorithm based on intrinsic motivation is proposed in this paper. The algorithm is applied to two-wheeled robot, which makes the robot gradually learn the corresponding relationship between behavior and reward or punishment signal, so that the robot can complete the self-balancing control.2) Learning automata algorithm based on intrinsic motivationAiming at the problem of the robot in unknown environment adaptive ability is poor,combined with psychology intrinsic motivation theory, put forward a kind of learning automata algorithm based on intrinsic motivation. The robot learns the corresponding relationship between the action and the reward signal, thus completing the path planning task.3) The robot cognitive developmental algorithm based on cerebellum-basal gangliacerebral cortex circuitsAiming at the problem of robot continuous behavior learning, according to the related knowledge of the sensory motor system, introduced the intrinsic motivation mechanism in psychology to drive robot autonomous learning, finally put forward the algorithm shows a good self-organizing, adaptive and self-learning ability.