Prediction Enhanced Immune Algorithm for Robot Path Plan and System Realization

• Main Authors: Kuan-ting Chou, 周冠廷
• Other Authors: Chih-Ming Chen
• Format: Others
• Language: zh-TW
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/9jyh62

碩士 === 國立臺灣科技大學 === 電機工程系 === 99 === Immune system is living body’s self-protection system. It takes the necessary defense and response measures by identifying invasive bodies from non-self substances, and by drawing from the past memories along with the regeneration of antibody. This adaptive-memory feature enables the emulation of immune system be applied to the optimal behavior decision making in dynamically changing environment, especially in the mobile robots path planning. This thesis develops an prediction enhanced rule for Ishiguro’s artificial immune algorithm by incorporating an estimation of the target position, and responds accordingly in the aforementioned technique. Computer simulation via MATLAB programming has proved the feasibility of this new approach. Hardware implementation, via Microchip Chip equipped mobile robot, further confirms that this new technique can be effectively applied in mobile robot obstacle avoidance and tracking tasks.