论文部分内容阅读
机器人在未知复杂环境下作业时,有从高处跌落的风险,所以需要考虑其空中的姿态调整能力,减轻由错误的着陆方式造成的伤害.当猫从高空坠落时,它总能通过自身调整安全落地.受这一生物学现象启发,通过研究猫能够安全落地的第一阶段,姿态调整阶段,首次提出以时间最优方式研究落猫机器人最优下落轨迹问题.首先利用轴对称双刚体模型导出机器人的数学模型,因为角速度不可积,将落猫机器人的姿态控制问题转变成非完整系统的运动规划问题.考虑到姿态调整所消耗的时间是决定调整结果较为重要的因素,随后以虚拟力矩输入取代真实角速度输入,建立时间优化函数,并构造了求解该函数的方法.然后以姿态调整耗时最短为控制目标,通过粒子群优化算法寻找目标最优解.最后运用所得最优解数据,在虚拟物理环境下进行了虚拟样机实验,并实现了落猫机器人的空中翻正动作.结果表明,该方法能有效地缩短自由下落机器人的姿态调整时间.
When a robot is working in an unknown complex, there is a risk of falling from a height, so its attitude in the air needs to be adjusted to reduce the damage caused by the wrong way of landing.When a cat crashes from height, it always adjusts itself This article first enlightens the optimal time to study the falling trajectory problem of the falling cat robot by enlightening this biological phenomenon by studying the first phase of the cat’s safe landing and the stage of attitude adjustment.Firstly, The mathematical model of the robot is derived because the angular velocity is not integrable, which will transform the attitude control problem of a cat into a motion planning problem of a nonholonomic system.Considering that the time consumed by attitude adjustment is a more important factor in determining the adjustment result, Inputting the input of real angular velocity instead of real angular velocity and establishing the time optimization function, and constructing a method to solve this function.Secondly, using the shortest time of attitude adjustment as the control target and finding the optimal solution by using particle swarm optimization algorithm.Finally, using the optimal solution data, In the virtual physical environment under the virtual prototype experiments, and to achieve a cat cat robot The righting operation. The results show that this method can effectively reduce the free fall of the robot attitude adjustment time.