Wenchuan Jia

Electroencephalography(EEG)-based instinctive brain-control of a quadruped locomotion robot

Artificial intelligence and bionic control have been applied in electroencephalography (EEG)-based robot system, to execute complex brain-control task. Nevertheless, due to technical limitations of the EEG decoding, the brain-computer interface (BCI) protocol is often complex, and the mapping between the EEG signal and the practical instructions lack of logic associated, which restrict the users actual use. This paper presents a strategy that can be used to control a quadruped locomotion robot by users instinctive action, based on five kinds of movement related neurophysiological signal. In actual use, the user drives or imagines the limbs/wrists action to generate EEG signal to adjust the real movement of the robot according to his/her own motor reflex of the robot locomotion. This method is easy for real use, as the user generates the brain-control signal through the instinctive reaction. By adopting the behavioral control of learning and evolution based on the proposed strategy, complex movement task may be realized by instinctive brain-control.

Generating Vectored Thrust With the Rotational Paddling Gait of an ePaddle-EGM Mechanism: Modeling and Experimental Verifications

An eccentric paddle mechanism based on the epicyclic gear mechanism (ePaddle-EGM), proposed to enhance the mobility of amphibious robots for multiterrains tasks, can execute several terrestrial and aquatic gaits. In this paper, we rectify the experimental setup as well as the thrust modeling to do further research on rotational paddling gait. The effects of the rotational period and the location of paddle shaft to the thrust are analyzed. Besides the amplitude features, experimental results also illustrate the vector character of thrust generated by the rotational paddling gait.

Mechanical design of a compact and dexterous quadruped robot

This manuscript introduced the modularized structural design of a medium sized quadruped robot with desired compact and dexterous mobility. In detail, the ball screw transmission and the four-bar linkages mechanism were adopted to construct the quadruped robot, and then the relationship between the torque of the knee joint and the torque of the motor was modelled and analyzed. Moreover, the mechanical performance of transmission is analyzed, then the practical result indicates that the shank should be configurated to get close as much as possible to the limited position of extension while walking. The loading analysis further showed that the structural strength of each leg is still enough, when adding 200N external force that similar to the whole weight of the robot to one foot. Based on the optimized design, the mechanical prototype was accomplished, and the corresponding control strategy were briefly illustrated. Finally, locomotory planning in typical walking and when the robot falls down were discussed preliminarily.

Stiffness analysis and verification of one 3 - RPS parallel sensor

Multi-axis force and torque measurement play an increasingly important role in robot technology. The stiffness, which is related to the mechanical structure and elastic material, is not only one of the fundamental characteristics of the FT sensor, but also the basis to calculate the dynamical performance of the sensor. In this manuscript, one type of 3-RPS parallel sensor is described, then the corresponding stiffness analysis and experiments are accomplished. The static model of the 3-RPS parallel sensor is established firstly, and the stiffness description of one single branch in parallel structure is illustrated. Then the overall stiffness model is presented, and some key parameters that influence the stiffness transfer matrix are discussed. The proposed model and calculation methods are analysed by FEM simulation. Finally, stiffness experiments are realized to verify and optimize the theoretical results.

Design and analysis of a transformable spherical robot for multi-mode locomotion

In this manuscript, a novel transformable spherical robot for multi-mode locomotion is proposed. The robot has the spherical shell and an internal driving unit for spherical rolling, and also has a quadruped unit for walking. The robot can change the locomotion mode between spherical rolling mode and the quadruped walking mode to adapt to varied terrains. Moreover, the analysis and simulation of the robot moving across the obstacle shows that hybrid driving mode has better mobile performance than basic driving mode. The experiments that the prototype moves in the real environment show that the robot could adapt more types of terrains.

The analysis and control method about the stop motion of Symmetric Planar Rimless Wheel on slope

Symmetric Planar Rimless Wheel (SPRW) is an important reference model of passive biped walking. Most of the researches on SPRW emphasis on the global characteristics but are lack of the detailed motion process. In this paper, the process of the stop motion of SPRW on slope and the relation between different kinds of motion form with SPRWs instinct characteristics are discussed. Furthermore, an approximate calculation method about the basin of attractions to different kinds of motion form is proposed. Finally, a control method about the stop motion of SPRW on slope is developed which is helpful to control complex Planar Rimless Wheel when it walks in the unknown environment.

Optimized non-reciprocating tripod gait for a hexapod robot with epicyclic-gear-based eccentric paddle mechanism

A novel eccentric paddle mechanism based on the epicyclic mechanism (ePaddle-EGM) has been proposed to enhance the mobility of amphibious robot for multi-environments tasks with diverse locomotion gaits, including a novel non-reciprocating legged gait. In this study, an optimized non-reciprocating planning method by planning the posture angle of the supporting paddle is focused to improve energetic efficiency of this gait. Relationship between the posture angle of the supporting paddle and actuation forces on the paddle is analyzed in the state of equilibrium. Standing on the ground vertically is found to be an optimal posture for the supporting paddle to achieve minimum quadratic sum of the actuation forces. The planning method that considers the optimal posture angle of the paddle and the stride of the gait is established and verified in simulations. Calculated specific resistance confirms that the proposed method can improve the energetic efficiency of the non-reciprocating legged gait.

Sliding mode control with adaptive feedforward compensator for ultra-precision active vibration isolation

Ultra-precision active vibration isolation system protects ultra-precision equipment from the transmission of external vibration, and suppresses vibration incurred by internal structures. In order to obtain the ideal skyhook effect, the sliding mode controller with adaptive feedforward compensator is proposed in this paper. With the controller, the vibration system can obtain the desire performance even when the system is under model uncertainties and significant direct disturbance. A series of simulations are carried out to verify the effectiveness of the controller.

Reliable control of a quadruped walking robot in uneven terrain environment based on noninvasive brain-computer interface

In this paper, a novel control architecture and strategies, involving noninvasive brain-computer interface (BCI), are presented, and a virtual prototype of quadruped walking robot, named QWR-I, is developed for simulation of navigation behaviors of the BCI-based robot. The BCI can provide limited patterns of users intention based on EEG signals, following users motor imagination. The proposed control architecture, dealing with both robots autonomous planning and users decision acquired from BCI, is elaborated. To achieve efficient navigation in uneven terrain environment where the robot is located, three control modes are proposed, taking trade-off between robots autonomy and users flexibility. Movement efficiency is priority in relatively even terrain by strolling mode, while safety is considered more in uneven terrain by terrain mode. The coding protocols, which semantically represent the same set of BCI signals with different meanings according to the situated control mode of the robot, and switching strategies between different modes are elucidated. To satisfy the users intention of desired path to destination to greatest extent possible on the premise of walking stability, cooperative control strategy based on centroid stability margin while the robot walks is presented.