Research On Multi-Sensor Data Fusion And Data Transmission Of Multi-Sensor Gripper

Space robots are installed sensors in their grippers so as to operate dexterously in an uncertainty environment. The multi-sensor data fusion technique can utilize sufficiently the outputs of sensors, process these data in different levels, and obtain the information of the robotic status and their working environments accurately and integrally. Therefore it plays the important role in the sensing system of robotic gripper. It is necessary for extravehicular mobile robot (EMR) adopting the multi-sensor data fusion technique to obtain the grasping information, such as the grasping force of gripper, the linking status between the gripper and I-shaped beam, and multi-dimension wrist force/torque. At the same time the data of sensors needs to be transmitted in EMR. The publication of the IEEE 1451 that is a standard of networked smart sensors provides the foundation for improving the transmitting data ways of sensors.The main contents in this dissertation include:1. According to outputs of eight finger force sensors the grasping force of gripper is obtained using the multi-sensor data fusion technique based on the BP artificial neural network.2.The status of grasping I-shaped beam with the gripper is also obtained using the multi-sensor data fusion technique based on the BP artificial neural network according to eight finger force sensors, four proximity sensors and a displacement sensor. It provides the information for ensuring the robot walking or grasping I-shaped beam safely and reliably.3.The estimation method of multi-dimension wrist force/torque is proposed, the project of experiment is designed, and experiments have been done. According to the output variations of finger force sensors, the multi-dimension wrist force/torque are estimated using the data fusion technique based on BP artificial neural network.4.The software of multi-sensor data fusion for EMR gripper is developed with Visual C++6.0. The information of multi-dimension wrist force/torque is displayed in real time during the robot works, which provides the foundation for the safe and reliable operations of robots.5.The question how to utility the IEEE 1451 standard is discussed, and the misapprehensions of relations among IEEE 1451.x Standards are clarified. The IEEE 1451.x standards may be together used to develop the networked smarted sensor, and can also be used to work in stand-alone mode.6.A networked smart sensor based on an embedded network module for EMR grippers is developed for data sampling, fusion and transmission. It is composed of the EMR gripper, the embedded network module and the data sampling circuit.7.A project of wireless networked smart sensor based on the embedded networked module is designed for robotic gripper.