Geometric design optimization of an under-actuated tendon-driven robotic gripper

• Published in: Robotics and computer-integrated manufacturing 2018-04, Vol.50, p.80-89
• Main Authors: Dong, Huixu, Asadi, Ehsan, Qiu, Chen, Dai, Jiansheng, Chen, I-Ming
• Format: Article
• Language: English
• Subjects: Algorithms; Design optimization; Fitness; Genetic algorithms; Geometric constraints; Geometric design optimization; Mandrels; Mathematical models; Pulleys; Robotic gripper; Robotic modelling; Robotics; Stability analysis; Tendon routes
• ISSN: 0736-5845; 1879-2537
• DOI: 10.1016/j.rcim.2017.09.012

•We build the mathematical model between the actuated force and the contact force for an under-actuated tendon-driven robotic gripper based on the geometric analysis.•A mathematical model is constructed to obtain the transmission efficiency of the tension force when a tendon wraps a joint mandrel by the geometric relations.•The geometric model of transmission characteristics determined by the tendon routes for reducing the resistance is generated.•Genetic Algorithm is applied to optimizing the dimensions of the gripper and the tendon routes.•The geometrically optimal approach provided by us has the characteristics of the versatility and can also be referred to optimizing most of the under-actuated robotic gripper with tendon-driven mechanisms. The design optimization of a robotic gripper is of utmost importance for achieving a stable grasp behaviour. This work focuses on analysing the optimal design of an under-actuated tendon-driven robotic gripper with two 3-phalange fingers and a geometric design optimization method is proposed to achieve a stable grasp performance. The problem has twenty-two design variables, including three phalange lengths, three phalange widths, three radii of joint mandrels, a palm width and twelve route variables for allocation of six pulleys. First, the mathematical model between the active and contact forces is expressed in relation to the geometric dimensions of the robotic gripper. Second, the geometric model of transmission characteristics determined by the tendon routes for reducing the resistance is generated. Next, three objective functions and multiple geometric constraints are derived and integrated into two fitness models. Finally, the genetic algorithm is applied to addressing the optimization problem. Practical experiments are performed as well to validate the proposed approach. The approach is universal for optimizing any conventional under-actuated tendon-driven gripper.