主催: 一般社団法人 日本機械学会
会議名: ロボティクス・メカトロニクス 講演会2020
開催日: 2020/05/27 - 2020/05/30
In recent years, the study of autonomous construction has been actively pursued. In construction work using hydraulic excavators, the sequence of excavation motions profoundly effects the duration and precision of work. To realize highly efficient, autonomous construction work, efficient planning of excavation motion sequences is required. However, there is currently no effective method for motion sequence evaluation, which prevents us from planning an efficient motion sequence. This research proposes an evaluation method of excavation motion sequences based on a symbol pattern analysis technique. In the proposed method, first, data array of hydraulic excavator motions are divided into individual excavation motions and a clustering method symbolizes the individual excavation motions. Second, symbol patterns are divided into excavation phases which have similar excavation motion patterns. Finally, excavation sequence is evaluated by calculating the frequency of the transition between excavation phases.