Study On Deep Seabed Mining Robot Vehicle Motion Modeling And Control

Deep seabed contains abundant mineral resource. Research on it's explore technique has important significance for our country's mineral resource continual owner and development of deep-sea resource exploitation technology. Deep seabed mining robot vehicle works in 6000m-depth seabed unknown complex environment. The control quality has direct influence on our countries ocean exploitation strategy. So, in the support of project named as Internal seabed area research and exploitation, the thesis proposes a movement modeling and control strategy for deep seabed mining robot vehicle. The main research achievements include:(1)Deep seabed mining robot vehicle movement modeling techniqueDeep seabed mining robot vehicle works in 6000m-depth extra-soft seabed, the special design and special working environment decides it's special working property. Aiming at the properties of high-sharp triangle teeth, large sink, high slip rate and low speed., after special consideration on track teeth thrust, bulldozer resistance, hydraulic resistance, and omitting centrifugal force, the dynamic model of deep seabed robot vehicle is build. After that, the kinematical model is built in Cartesian coordinate system with consideration of track slip rates.Aiming at properties of complex parameters and high nonlinearity for the vehicle's variable-hydraulic pump to fix-hydraulic motor volume velocity modulation system, the system is composed as electric-hydraulic proportional directional valve, variable-pump-controlled cylinder, piston pump and piston motor module to build model separately. On the basis of the four sub modules, deep seabed mining robot vehicle hydraulic drive model is built.Combining the dynamic model, the kinematics' model and the hydraulic model, and realized in Matlab language, he deep seabed mining robot vehicle movement system simulation model is finished. Some simulations are done to verify the validity of the model.(2) Deep seabed mining robot vehicle key moving parameters identification techniqueBecause of the unknown mining environment, uneven and extra soft seabed sediment, the vehicle works with high slip and big uncertain moving state change. However, the key moving parameters such as effective radius of driving wheel and track slips are difficult to measure. A new approach identify these parameters is presented. First, based on moving analysis, the parameters identification model isbuilt. The model is used to online calculate slip rate and effective radiuses of driving wheels by measurement on the motors' pressure and the sediments. After that, selecting several state variables, the nonlinear parameters estimation model of deep seabed mining robot vehicle's left and right track slip rates and effective radiuses of driving wheels is built. It builds basis for optimal unbiased estimation of deep seabed mining robot vehicle's key moving parameters.Then, an improved SUKF algorithm-FSUKF is presented: fuzzy algorithm is used to regulate sigma set operator, which makes ideal model be more identify to real model. The Mackey -Glass time series model is used to verify the validity of the algorithm.In the end, the proposed model and algorithm are used to estimate the key parameters; the simulation result verifies the validity of the method.(3)Deep seabed mining robot vehicle motion control techniqueBased on deep seabed mining robot vehicle control hardware structure and working rule, a motion control system of the vehicle is presented. The system is consisted of modules such as movement planning, parameter estimation, trajectory tracking error computation and tracking control etc. Functions of the Modules are designed separately.Usual sampling approach for trajectory planning is equal interval sampling. In order to improve control precision, a sample approach based on fuzzy rule is presented. This approach has the function of on-line tuning distance between vehicle's real place and target, according to the distance and angle errors.After analysis of tracked vehicle's tracking errors, an expert-fuzzy based cross coupling controller is designed. The internal error is deal with cross-coupling controller; an expert-fuzzy controller eliminates the external error. Simulation results are provided to verify the proposed scheme.(4) Deep seabed mining robot vehicle simulation system developmentDeep seabed mining robot vehicle is voluminous and needs high driving power. It's too expensive for usual motion control exercise. So, a small deep seabed mining robot vehicle model and a simulation system is developed to solve the problem. Some exercise is done based on vehicle model, the result is provided also.