| Humans have never ceased their thirst for human enhancement since ancient times,and the improvement of load-bearing capacity has certain practical significance.Nowadays,there are less work depending on human body with the development of technology.However,there are still many jobs that require the human body to carry a certain amount of weight.For example,soldiers have to carry heavy equipment on the battlefield and workers have to carry heavy tools in the factory.The heavy load on human body in this kind of work not only affects the work efficiency,consumes a lot of physical strength,but also may cause damage to the human body.To solve this problem,a large number of researchers at home and abroad have focused their attention on wearable lower-limb assistive exoskeleton robots for load carrying,and a certain number of research results have been produced.However,most of the reported lower-limb assistive exoskeleton prototypes have limited load capacity,and need to be adjusted manually for wearers of different heights.Besides,further research on human-machine compliant interaction and human-machine coordination control is needed.In this paper,a parallel-configuration lower limbs powered exoskeleton robot for high-load carrying is proposed.The system design,kinematics and dynamics analysis,force control of the drive unit,and humanmachine coordinated control strategy of the exoskeleton robot are carried out.Finally,the performance of the developed exoskeleton robot is verified by experiments.Based on the physiological structure and motion requirements of the human body lower limbs,the characteristics of exoskeleton robots that are homogeneous and heterogeneous with the lower limbs of the human body are analyzed.The non-anthropomorphic parallel configuration of the exoskeleton leg is selected,and the degree of freedom configuration and the form of the drive joints are determined.The sizes of leg structural link are optimized based on genetic algorithm.The mechanical systems,including the back,the leg with parallel structure and the shoes with force measurement function,are designed based on design indicators and requirements.To improve the load carrying capacity of the drive unit and achieve compliance,a series-parallel elastic actuator is designed based on the output characteristics of leg when the human body is walking.Considering the requirements of exoskeleton mechanical system and control strategy,a lightweight and integrated electrical system based on CAN bus is designed.Finally,a parallel-configuration lower limbs assistive exoskeleton prototype with a self-weight of 27.8kg is developed.To analyze the structural characteristics of the exoskeleton,the kinematic model of the simplified leg parallel structure is established,and the workspace,singularity,velocity and static transfer characteristics are analyzed.The kinematic model of the exoskeleton system is established using D-H coordinate system.Taking walking as an example,the human movement process is analyzed,and the motion phases of legs are divided into stance phase and swing phase.A unified dynamic model of the exoskeleton is established with floating base without distinguishing the motion phases,and the required force of the drive unit under different motion states is calculated,which lays the foundation for subsequent research on control strategy of the exoskeleton.Force control of the drive unit is the basis for the upper-level motion control of the exoskeleton.Combining with dynamic compensation,feedback control and disturbance observer,a force control method for the exoskeleton series-parallel elastic actuator is designed.The parameters of Lu Gre friction model are identified by genetic algorithm,and the dynamic compensation controller of drive unit is obtained combining with force feedback information at the end of the actuator.Cascade control based on velocity inner loop is used as feedback controller,and a nonlinear disturbance observer is used to suppress the influence of unknown disturbances on the control system.Simulations and experiments are carried out to verify the effectiveness of the force control method.The designed force controller of the drive unit can make the drive unit respond to the step control command with the size of 400 N within 120 ms,and adapt to the different stiffness of the series elastic module and impedance at the load end.A dual-mode motion coordination control method is designed,and a compensation control method based on dynamic model is adopted in non-walking mode.The relationship during movement between human and exoskeleton robot is analyzed,and a human-machine coordinated walking control strategy based on central pattern generator is proposed.The central pattern generator network structure is established according to the requirements of the exoskeleton.The network parameters are optimized offline with genetic algorithm and dynamic simulation software,and the optimal reference motion trajectory is generated by the central pattern generator.Impedance control of the drive unit is introduced to meet the requirements of flexibility and comfort during walking.The parameters of impedance control,the period,amplitude and phase of the reference trajectory are adjusted online according to the actual motion information.The effectiveness of trajectory online adjustment method is verified by experiments.The function and performance evaluation experiments of the developed parallelconfiguration lower limbs assistive exoskeleton robot system are carried out.The flexibility and adaptability to the terrain are verified through several experiments with different movements and different terrains.The performance indicators of the exoskeleton are obtained through the walking speed experiment,squatting and walking experiments with different loads.The results show that the developed exoskeleton has a good assist effect in the process of squatting and walking.The maximum walking speed of the exoskeleton is 3.6km/h,and the maximum load is 88 kg. |
PDF Full Text Request