并联协作机构轨迹规划研究

发布时间：2018-12-18 19:26

【摘要】：目前国内外在并联协作机构轨迹规划领域的相关研究报导很少，本文作为一个探索性课题，基于两个已知参数的Stewart机构构建协作仿真平台，分析协作机构的运动学逆解与工作空间并研究并联协作机构轨迹规划整个过程。考虑协作曲面的任意性，本文采用NURBS对协作曲面进行表示，并分析基于NURBS表示的协作曲面造型与实时插补。造型的难点在于NURBS曲面的裁剪，而曲面裁剪的核心在于边界曲线三维坐标空间在NURBS曲面uv参数空间的投影，本文设计一种基于牛顿迭代的NURBS曲面快速裁剪方法，实现了课题NURBS曲面的有效裁剪问题。插补方面，本文设计一种基于自调节等参数线法的u向步长求取结合基于自适应速度的v向步长求取的协作机构实时插补方法。设计协作优化模型进行并联协作机构的协作轨迹规划分析。协作优化模型可描述为一个基于三优化参数的时变协作任务分解与分配模型，即采用一组优化参数对时变协作任务进行分解与分配，将协作任务的分解与分配问题转化为一个最优化问题。针对并联协作机构位姿（包括位置信息和姿态角信息），本文完成了并联协作机构轨迹规划过程中的位置协作，即对协作机构平移矩阵进行优化分解与分配。根据NURBS的协作曲面造型，实时插补算法和协作优化模型，基于Linux系统，采用GTK+和OpenGL编程进行协作轨迹规划仿真软件开发，，实现实时显示并联协作机构协作轨迹规划过程，同时验证本文设计的协作优化模型和协作轨迹规划算法的可行性与正确性。
[Abstract]:At present, there are few reports in the field of parallel cooperative mechanism trajectory planning at home and abroad. As an exploratory topic, this paper constructs a collaborative simulation platform based on two known parameters of Stewart mechanism. The kinematics inverse solution and workspace of cooperative mechanism are analyzed and the whole process of trajectory planning of parallel cooperative mechanism is studied. Considering the arbitrariness of the cooperative surface, this paper uses NURBS to represent the cooperative surface, and analyzes the modeling and real-time interpolation of the cooperative surface based on the NURBS representation. The difficulty of modeling is the cutting of NURBS surface, and the core of surface clipping is the projection of 3D coordinate space of boundary curve in uv parameter space of NURBS surface. In this paper, a fast clipping method of NURBS surface based on Newton iteration is designed. The effective clipping problem of NURBS surface is realized. In terms of interpolation, this paper designs a real-time interpolation method for cooperative mechanism based on self-adjusting isoparametric line method, which combines u direction step size with v direction step size based on adaptive speed. The collaborative trajectory planning analysis of parallel cooperative mechanism is carried out by designing a collaborative optimization model. The cooperative optimization model can be described as a time-varying cooperative task decomposition and assignment model based on three optimization parameters, that is, a set of optimization parameters is used to decompose and allocate the time-varying cooperative task. The problem of decomposition and assignment of cooperative tasks is transformed into an optimization problem. Aiming at the position and pose of parallel cooperative mechanism (including position information and attitude angle information), this paper completes the position cooperation in the process of parallel cooperative mechanism trajectory planning, that is, the optimal decomposition and allocation of the translation matrix of the cooperative mechanism. According to NURBS's collaborative surface modeling, real-time interpolation algorithm and cooperative optimization model, based on Linux system, the collaborative trajectory planning simulation software is developed by using GTK and OpenGL programming, and the collaborative trajectory planning process of parallel cooperative mechanism can be displayed in real time. At the same time, the feasibility and correctness of the collaborative optimization model and the collaborative trajectory planning algorithm designed in this paper are verified.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH112

【参考文献】