Browsing by Issue Date, starting with "1999-07-21"
Now showing 1 - 3 of 3
Results Per Page
Sort Options
- A trajectory planner for manipulators using genetic algorithmsPublication . Pires, E. J. Solteiro; Tenreiro Machado, J. A.This paper proposes a genetic algorithm (GA) to generate trajectories for robotic manipulators. The objective is to minimize the ripple in the time evolution of robot positions and velocities. Moreover, the manipulator is required to reach a predefined goal without colliding with obstacles in the workspace. The article presents the results for several redundant and non-redundant robot manipulators.
- A trajectory planning algorithm for redundant manipulatorsPublication . Duarte, Fernando B. M.; Machado, J. A. TenreiroRedundant manipulators have some advantages when compared with classical arms because they allow the trajectory optimization, both on the free space and on the presence of obstacles, and the resolution of singularities. For this type of manipulators, several kinematic control algorithms adopt generalized inverse matrices. In this line of thought, the generalized inverse control scheme is tested throught several experiments that reveal the difficulties that often arise. Motivated by these problems this paper presents a new method, that optimizes the manipulability through a least square rational function approximation, to determinate the joints positions and studies the chaos revealed by the kinematics trajectory planning scheme, as well as its influence on the dynamics, when controlling redundant and hyper-redundant manipulators. The experiments confirm the superior performance of the proposed algorithm, for redundant and hyper-redundant manipulators, revealing several properties and gives a deeper insight towards the future development of superior trajectory control algorithms.
- Fourier analysis of robot trajectories in random tasksPublication . Tenreiro Machado, J. A.; Galhano, AlexandraA new method for the study and optimization of manipulator trajectories is developed. The novel feature resides on the modeling formulation. Standard system descriptions are based on a set of differential equations which, in general, require laborious computations and may be difficult to analyze. Moreover, the derived algorithms are suited to "deterministic" tasks, such as those appearing in a repetitive work, and are not well adapted to a "random" operation that occurs in intelligent systems interacting with a non-structural and changing environment. These facts motivate the development of alternative models based on distinct concepts. The proposed embedding of statistics and Fourier transform gives a new perspective towards the calculation and optimization of the robot trajectories in manipulating tasks.