This paper proposes a method that automatically designs the sensory morphology of a mobile robot. The proposed method employs two types of adaptations -ontogenetic and phylogenetic- to optimize the sensory morphology of the robot. In ontogenetic adaptation, reinforcement learning searches for the optimal policy which is highly dependent on the sensory morphology. In phylogenetic adaptation, a genetic algorithm is used to select morphologies with which the robot can learn tasks faster. Our proposed method is applied to the design of the sensory morphology of a line-following robot. We carried out simulation experiments to compare the design solution with a hand-coded robot. The results of the experiments revealed that our robot outperforms a hand-coded robot in terms of the line-following accuracy and learning speed, although our robot has fewer sensors than the hand-coded one. We also manufactured a physical robot using the design solution. The experimental results revealed that this physical robot uses its morphology effectively and outperforms the hand-coded robot.
|Journal||IEEJ Transactions on Electronics, Information and Systems|
|Publication status||Published - 2008|
- Ecological balance
- Learning and evolution
- Sensor evolution
ASJC Scopus subject areas
- Electrical and Electronic Engineering