A novel trajectory planning approach with torque and jerk constraints based on polynomial interpolation profile and adaptive iteration

This paper presents a novel adaptive iteration approach to trajectory planning for industrial robots, with bounded torque and jerk. The core of this approach is to construct a conservative velocity curve and iteratively increase the feasible velocity curve while adhering to kinematic and dynamic constraints. In this approach, a polynomial profile is given to achieve smooth velocity transitions between adjacent path points, effectively modeling the acceleration and deceleration processes. Based on this polynomial profile, a conservative velocity curve consisting of three stages—acceleration, constant velocity, and deceleration—is constructed. The velocity in the constant velocity stage and the number of path points involved in the acceleration and deceleration stages are determined using the bisection method. Subsequently, the velocity in the constant velocity stage of the conservative or the previous velocity curve is increased following the same way that constructs the conservative velocity curve. This process is repeated until the number of path points in all constant velocity stages is below a given threshold. The proposed approach can be implemented on complex geometric paths of a 6-DOF manipulator while satisfying all kinematic and dynamic constraints. Compared to the comparison method based on convex optimization, the proposed method can reduce the traversing time by 6.06 % and the computation time by 77.4 %.