Abstract This paper proposes a methodology to generate online smooth joint trajectories of robots based on an improved sinusoidal jerk model. The multi-segment trajectory model is designed to allow a more sufficient exploitation of the actuation capability, thereby shortening the execution time while ensuring continuity up to the jerk level. Afterwards, the general mathematical expressions of the motion profiles are derived from integration of the jerk function. By analyzing the critical constraint conditions corresponding to each possible profile type, a complete closed-form solution to the minimum time planning problem with consideration of kinematic constraints has then been worked out, where a parameter named the ramp coefficient that manages the ratio of the sine jerk period is introduced to simplify the determination of the motion formulas. Thus the method results in a computationally light optimization procedure for facile implementation and enables a higher level of regularity for achieving desirable motion behaviours in light of the operation requirements and joint characteristics. Experimental results conducted on a robot manipulator reveal that the proposed approach is capable of yielding better performance than existing techniques in terms of efficiency and jerk suppression.

中文翻译：

---

一种具有运动学约束的机器人机械手的急动连续轨迹生成方法

摘要 本文提出了一种基于改进的正弦加加速度模型生成机器人在线平滑关节轨迹的方法。多段轨迹模型旨在允许更充分地利用驱动能力，从而缩短执行时间，同时确保连续性达到加加速度水平。之后，运动曲线的一般数学表达式是从加加速度函数的积分中推导出来的。通过分析与每种可能的轮廓类型相对应的临界约束条件，然后制定了考虑运动学约束的最小时间规划问题的完整封闭形式解决方案，其中一个名为斜坡系数的参数管理正弦比引入急动周期以简化运动公式的确定。因此，根据操作要求和关节特性，该方法导致计算量轻的优化程序以易于实施，并且能够实现更高水平的规律性以实现期望的运动行为。在机器人操纵器上进行的实验结果表明，所提出的方法在效率和抖动抑制方面能够产生比现有技术更好的性能。