Complementary integration framework for localization and recognition of a humanoid robot based on task-oriented frequency and accuracy requirements

摘要: A robot system that can process environmental measurements and motion planning during locomotion is necessary to continuously perform various tasks. To achieve such a system, which we call the Perception-during-Traversing Model, accuracy of recognition must be improved computational costs reduced; these are tradeoff relationships. In this paper, propose construction framework for humanoid solve trade-off problems Model system. The key idea proposed subdividing re-integrating localization processes in complementary manner based on task-oriented frequency requirements. Moreover, apply our JAXON, demonstrate it execute tasks by Model. most important contribution enabling localize itself accurately measure environment densely enough using its on-board computers; provides practical solution between quality costs.