Estimation Of Motion Parameters And Coordinated Tracking Control Of Planetary Rover On Soft Uneven Terrain

The Chang'e IV landed on the back of the moon in 2019.It represents the remarkable achievements of China in lunar exploration.In addition,with the continuous advancement of Mars exploration in recent years,planetary exploration has returned to the public's vision.As the platform of planetary exploration mission,the planetary rover plays an important role in the mission.Considering the actual situation of the soft soil and rugged terrain on the planet surface,how to control the motion of the planet vehicle in this case becomes the key problem to complete the scientific exploration mission safely and efficiently.At present,most of the research on planetary vehicles and wheeled mobile robots still satisfy the condition of non-holonomic constraints,and the terrain is flat terrain.The existence of slip phenomena leads to the destruction of non-holonomic constraints and the performance of the control laws on flat terrain cannot be applied on rugged terrain directly,which challenges the trajectory tracking control of planetary rover.In addition,the planetary rover generally has more than two wheels and each wheel drives independently,which makes it a typical redundant driving system.How to optimize and coordinate the control input of each wheel to reduce unnecessary internal force loss is also a worthy problem.This paper takes the six-wheeled rocker-bogie planetary rover as the research object.To solve two kinds of problems above,considering the slip/slip and redundant driving,the trajectory tracking control on the soft and rugged terrain and the optimal coordinated allocation of driving wheels of the planetary rover are studied.For most of the current research on tracking control of wheeled mobile robots,kinematics/dynamics modeling is based on two-dimensional plane.In this paper,a three-dimensional kinematics and dynamics models of planetary rover on soft and rugged terrain are established.At the same time,according to the controllable degree of freedom in actual operation,the models are simplified.According to the relationship between the joints of the planetary rover suspension,the Jacobian matrix between the rover velocity and the wheel velocity and the transformation relationship between the rover coordinate system and the wheel axle central coordinate system are deduced.The velocity of each wheel can be calculated according to the rover velocity,and the force of the rover can be calculated according to the forces of wheels.At the same time,the key parameters,the wheel-ground contact angle and the slip ratio,which need to be acquired in the modeling,are estimated to be applied to the following tracking control and coordinated control.The validity of the key parameter estimation method is verified by simulation in ROSTDyn.On the soft and rugged terrain,the slip phenomenon will affect the tracking of the desired trajectory.At the same time,the planetary rover system belongs to a strong coupling,non-linear multi-input and multi-output system,and it is relatively difficult to design the control law.To solve these two problems,based on the model proposed above,an improved ADRC controller is designed to track the desired trajectory,which can decouple the MIMO system,estimate and compensate the total disturbance in real time.The planetary rover has four independent steering wheels.In trajectory tracking,it is necessary to adjust the steering angles of four wheels to improve the tracking performance.Considering the influence of slip on the motion of the planetary rover,the steering control is studied,and the calculation formulas of steering wheel angle are obtained.The effectiveness of the proposed control law is verified by tracking the cosine curve and circular curve in the simulation platform.The multi-wheeled independent-driving planetary rover belongs to the redundant driving system,and the number of control inputs is larger than the controlled degree of freedom.Therefore,coordinated allocation of the control input of each wheel is worth studying to reduce the internal force loss caused by different motion states on soft and rugged terrain.To solve this problem,a coordinated control method considering both force and motion information is proposed.The control inputs considering motion information are obtained by the trajectory tracking method based on kinematics model,and the control inputs considering force information are obtained by the H₂/H?-QP controller based on dynamic model.By weighting the two control inputs,the control inputs of coordinated tracking control are finally obtained.The effectiveness of the control law is verified by simulations.Finally,the experimental platform architecture is improved,and the algorithm is transplanted.A sand field is built to simulate the soft and rugged planetary soil environment,and the visual capture system is included.A series of rover experiments verify the effectiveness of the tracking control and coordinated allocation algorithm studied above.