CAJ | 학술논문

机器人在对于一些气味源、声源、平衡位置等目标搜索时,为了精确定位,一般采用变步长的算法.根据S函数,设计出了一种新型步长调整算法.与传统的最陡下降法相比,其步长调整速度快,适应外部环境能力强,并且包含了方向信息.实验证明该方法是一种有效的移动机器人步长调整算法.
궤기인재대우일사기미원、성원、평형위치등목표수색시,위료정학정위,일반채용변보장적산법.근거S함수,설계출료일충신형보장조정산법.여전통적최두하강법상비,기보장조정속도쾌,괄응외부배경능력강,병차포함료방향신식.실험증명해방법시일충유효적이동궤기인보장조정산법.
Variable step-size algorithm is used when the mobile robots work at precise odor or sound source localization tasks.A novel variable step-size algorithm based on the Sigmoid function is proposed.Compared with the traditional steepest descent algorithm,the new algorithm not only modifies the step-size much faster,but also includes the direction information.The result of the experiments shows the efficiency of the new algorithm.