CAJ | 학술논문

在足球机器人系统路径规划问题的研究中，足球机器人系统工作的环境很复杂，既要配合本方机器人协同作战，还要对抗敌方机器人。针对传统的机器人路径规划算法过于复杂，同时没有充分考虑到足球机器人在比赛中实时性和对抗性，导致实时性差以及射门准确率低等问题，提出了一种在基本微粒群算法中加入交叉算子和变异算子的改进微粒群算法。种群中的所有微粒所经历过的最佳位置为当前全局最优位置。每个微粒都记忆当前自己所经历过的最优位置，将自己的最佳位置和全局最佳位置进行比较分析，修正自己的位置和速度。仿真结果显示改进微粒群算法比基本微粒群算法具有更优秀的搜索能力，实现了更快的收敛速度和更高的成功率，且对环境的变化有一定的适应性。
재족구궤기인계통로경규화문제적연구중，족구궤기인계통공작적배경흔복잡，기요배합본방궤기인협동작전，환요대항활방궤기인。침대전통적궤기인로경규화산법과우복잡，동시몰유충분고필도족구궤기인재비새중실시성화대항성，도치실시성차이급사문준학솔저등문제，제출료일충재기본미립군산법중가입교차산자화변이산자적개진미립군산법。충군중적소유미립소경력과적최가위치위당전전국최우위치。매개미립도기억당전자기소경력과적최우위치，장자기적최가위치화전국최가위치진행비교분석，수정자기적위치화속도。방진결과현시개진미립군산법비기본미립군산법구유경우수적수색능력，실현료경쾌적수렴속도화경고적성공솔，차대배경적변화유일정적괄응성。
Working environment of the robot soccer system is very complex. It is necessary to cooperate with teammates and fight against the opposite robots. Traditional robot path planning algorithms were complex and not fully took into account real-time in the game of soccer robot and confrontational issues, leading to poor real-time and low shooting accurate. Crossover operator and mutation operator were added in the basic particle swarm algorithm in the paper. All the particles in the population have experienced the best position as the current global optimal position. Each particle remember experienced the optimal position of their own, compared the best position with the global best position, and modified their position and speed. Simulation results show that this algorithm is correct and effective with better search ability, fast convergence speed, higher success rate and good environmental adaptation.