This work presents multi-functional robot arm gripper design along with vision and tactile sensor for efficient grasping and manipulation tasks. The design emulates human’s hand fingers structure using linkages and direct drive through slider-crank mechanism transmission. The structural elements are optimized for a finest performance in motion and force transmissibility of the gripper fingers. The main future of this design is its reliability to grasp and manipulate unknown object while its system complexity is reduced. The gripper has a tool change fixture incorporated into its palm, which will reduce time wastage and do assembling in one go. The gripper is equipped with two cameras in its palm; subsequently it will efficiently seek the target object and perform its prehensile task with intelligently determined grasping force.

For most quadruped robots, their waist joints can usually pitch or yaw, but cannot roll. Thus, their gaits can only be simulated by a simple motion model based on single-legged mechanisms. When pandas move on their four feet, they swing their hips and rear legs from side to side. Thus, the purpose of this study is to develop a quadruped robot which is equipped with a waist joint of one degree of freedom (DOF) for rolling so as to imitate the waist-swinging motion of a real panda. This research starts at editing the predetermined motion conditions with lengths of all panda body parts and the mechanism model with corresponding degree-of-freedom in the CAE software. The related simulation parameters of leg motions are acquired as a reference for gait controls. The quadrupedal walking process was tested in simulators to verify important design variables and simulation validity was also verified on the actual panda robot after their buildup. With this approach, this study has effectively and successfully developed a panda robot with a waist joint that can roll.

The present study proposes a windshield warning system for vehicles that marks potential hazardous objects in front via marks on the windshield. The target is identified by the camera and the software system and then used to construct the line-of-sight equation based on the coordinate system of the moving vehicle. The explicit equation of the intersection point of the line of sight and the windshield surface is derived using an algebraic method. A warning mark is then projected at the intersection point on the windshield by a two-degree-of-freedom laser projector, allowing the driver to easily identify the obstacle. The feasibility of the proposed system was verified using an actual vehicle. Results show that the system can perform the required function in real time.

No abstract

No abstract