The purpose of this study is to propose a trajectory generation method for 6 degrees of freedom manipulators that simultaneously optimizes all joint motions for point to point and multipoint tasks. Conventional methods treat each joint independently and synchronize to the slowest, often leading to suboptimal execution time and motion smoothness. The goal is to improve time efficiency and dynamic performance in industrial robotics by leveraging joint-level kinematic constraints.
A nonlinear optimization problem is formulated using quintic B-spline interpolation to generate smooth joint trajectories under velocity, acceleration and jerk limits. Segment durations for all joints are optimized simultaneously. The interior-point method is used to solve the problem. The approach is validated through simulation and physical execution on the Dobot Nova5 manipulator using benchmark point-to-point and multipoint motion tasks.
The proposed method yields trajectories up to 39% faster than conventional methods in point-to-point tasks. It produces smoother motions with lower peak and average jerk values, improving execution quality. In multipoint scenarios, the method reduces execution time by 3.26% compared to the synchronous B-spline method. Both simulation and experimental results confirm the method’s effectiveness and accuracy. Comparative analysis with recent trajectory planning methods further demonstrates the proposed method’s superior C2 continuity and smooth motion characteristics.
The proposed trajectory optimization method considers only kinematic constraints, excluding dynamic factors such as torque limits, payload effects and actuator saturation. The framework assumes offline planning under fixed task conditions, limiting adaptability in dynamic environments. Obstacle avoidance is handled via predefined waypoints rather than integrated constraints. Although quintic B-splines offer a balance between smoothness and computational efficiency, alternative spline orders may improve performance in specific contexts. Future work will address dynamic modeling, real-time replanning and integration with collision-aware path planning to enhance practical applicability.
The proposed method enables smooth, time-optimal trajectory generation for robotic manipulators without synchronizing all joints to the slowest one, improving motion efficiency and productivity. Its asynchronous formulation allows better utilization of joint capabilities under kinematic constraints. By supporting both point-to-point and multipoint tasks, the approach is applicable to a wide range of industrial assembly operations. The method has been validated through hardware experiments, demonstrating consistent execution and repeatability. It can be readily integrated with existing robot controllers and extended with via point-based obstacle avoidance for constrained environments.
The proposed trajectory planning method enhances the operational efficiency of robotic manipulators, contributing to increased automation in manufacturing and assembly tasks. By enabling smoother, faster and more reliable robot motion, it supports safer and more collaborative work environments, especially in human−robot shared workspaces. Improved efficiency and accuracy in industrial processes may reduce energy consumption, material waste and production costs. This advancement supports broader goals of sustainable automation and makes advanced robotics more accessible for small and medium enterprises, potentially impacting workforce roles and productivity in evolving industrial landscapes.
Unlike traditional decoupled methods, this unified framework globally optimizes all joint motions using quintic B-splines under kinematic constraints. The method is adaptable across robotic platforms and suitable for computer numerical control feedrate planning. It enables smooth and time-efficient trajectories for advanced industrial applications.
