A Rearch On Joint Control Of Sampling Robotic Arm With Lightweight And Its Trajectory Planning

The mass and power consumption are two important indexes of planetary rover. The mass determines the launch capability and launch costs of the space exploration, while the power consumption determines how many scientific instruments it can equip with and how long they can continuingly work. The sampling robotic arm is an important component of the planetary rover, and it has some impacts on the performance of the whole planetary rover. So the researches of it have a significance role. A sampling robotic arm with small mass, low power consumption, compact structure, high reliability and simple control strategy was aimed to developed. This article have made a deep research on the joint system and track planning of it.First of all, a scheme which the sampling manipulator has 4 DOF with light weight, low power consumption and multifunction was proposed. In this proposal, the manipulator can not only collect and take back the samples, but also support the navigation camera as the supporting mast. A locking device is designed to form a triangular structure which is stable without any energy consumption. To achieve the compactness, reliability and low control error goals, a double-side-encoder for the joint was designed, where each joint equipped the incremental encoder at the input terminal of the motor and the absolute position encoder at the output terminal of the connecting bar instead of brake and torque sensor. Not only can this proposal reduce the complexity, but also be compact and reliable. And a control system composed of active position mode and passive friction compensation mode was designed to control the manipulator. Through pre-designed control sequence, it can complete various scientific tasks commendably.Furthermore, the MDH method was used to establish coordinate system of the manipulator, and the kinematics and dynamics equations were deduced, then the dynamic characteristics were analysed in MATLAB. Based on kinematics analysis, the trajectory planning in joint space and Cartesian Space were researched. In Joint Space, cubic polynomial, quintic polynomial and parabola fitting with linear interpolation are studied, and the solution equation is derived. In Cartesian Space, spatial linear interpolation and arc interpolation are studied and the interpolation process and program code were provided for its simulation. According to the proposed joint transmission, the transfer function of the joint was established, and the velocity loop and position loop of a single joint were simulated by using MATLAB/Simulink.Then, two model-based control methods including torque feedforward and torque computing were used to compensate for various nonlinear interference. The simulation results show that the response error was less than 10-4.The co-simulation system was builded with MATLAB and ADAMS software, and the working mode of sample acqusition and sample transfer were simulated. The simulation results of the position, speed, acceleration and torque curves of the joint lay a foundation for the future debugging of physical prototype.Last but not least, the physical prototype was made by simplifing joint structure, the hardware and software system of the manipulator were built and PC control software was developed in VC++ platform. A series of job tasks such as sample acqusition and transfer, the mass as well as the power consumption were tested on sample manipulator.