Sergei Savin

Parameter Optimization for Exoskeleton Control System Using Sobol Sequences

The focus of this paper is the control system of an exoskeleton that performs sit-to-stand motion. In previous publications it was shown that during such motion an exoskeleton can be modeled as a four bar serial mechanisms. That allows to simplify the control system design, which has been shown in the literature. This work provides further development of one of the existing approaches in designing control systems for exoskeletons performing sit-to-stand motion. In the paper a method for parameter optimization of the regulator is presented. The method is based on a multi stage procedure and combines the use of Sobol sequences with a nonlinear numerical optimization techniques. The results of the optimization and their analysis are presented. Relative advantages of using different objective functions are discussed.

Adaptive control system for exoskeleton performing sit-to-stand motion

In this paper, an exoskeleton performing a sit-to-stand motion is studied. A new approach to control system design for this mechanism is proposed. Proposed control system uses information about currents in the motors armature cores to assess how much the system is loaded and change its behavior accordingly. It also uses ZMP (zero moment point) principle for assessing systems balance. In the paper, some results of the numerical simulation are presented. On a few examples it was shown how changing duration of sit-to-stand motion influences dynamical behavior of the mechanical system.

Improvement of energy consumption for a lower limb exoskeleton through verticalization time optimization

This paper focuses on a lower limb exoskeleton. The problem of improvement of energy efficiency of the exoskeleton during sit-to-stand motion (verticalization) is considered. Optimization of the time allocated for the verticalization motion is proposed as a way to improve energy efficiency. It is shown that optimal time of verticalization depends on the initial position of the exoskeleton and the relation between the two can be approximated by a polynomial function. An analysis and suggestions for practical applications of the obtained results are presented.

Footstep Planner Algorithm for a Lower Limb Exoskeleton Climbing Stairs

In this paper, a footstep planning algorithm for a lower limb exoskeleton climbing stairs is presented. The algorithm relies on having a height map of the environment, and uses two procedures: partial decomposition of the supporting surface into convex obstacle-free regions, and optimization of the foot step position implemented as a quadratic program. These two methods are discussed in detail in the paper, and the simulation results are shown. It is demonstrated that the algorithm works for different staircases, and even for the staircases with obstacles on them.

Footstep planning for a six-legged in-pipe robot moving in spatially curved pipes

This paper presents a footstep planning algorithm for a six-legged in-pipe robot moving in spatially curved pipes. The algorithm allows us to generate sequences of points on the pipes inner surface where the contact pads of the robot should come in contact with the surface. The algorithm uses mapping of the pipe onto a two-dimensional surface with a simpler geometry. The algorithm plans footstep sequences on that two-dimensional surface and then maps them back onto the pipe. It is shown that this procedure can be extended by implementing an obstacle avoidance algorithm formulated as a quadratic program.

Motion Control Algorithm for a Lower Limb Exoskeleton Based on Iterative LQR and ZMP Method for Trajectory Generation

In this paper a problem of controlling a lower limb exoskeleton during sit-to-stand motion (verticalization) in sagittal plane is studied. It is assumed that left and right sides of the exoskeleton are moving symmetrically. The main challenge in performing this motion is to maintain balance of the system. In this paper we use the zero-moment point (ZMP) methodology to produce desired trajectories for the generalized coordinates that would allow the system to remain vertically balanced. The limitations of this approach is that, it requires relatively accurate work of the feedback controller that ensures that the exoskeleton follows generated trajectories. In this work we use Iterative Linear Quadratic Regulator (ILQR) as a feedback controller in order to obtained the required accuracy. In the paper a way of trajectory generation that uses ZMP methodology is discussed, the results of the numerical simulation of the exoskeleton motion are presented and analyzed. A comparison between a natural human motion (for a human not wearing an exoskeleton) and the simulated motion of an exoskeleton using the proposed algorithm is presented.

Motion Control Algorithm for Exoskeleton Push Recovery in the Frontal Plane

In this paper a full body assistive exoskeleton is considered. A mathematical model for the case of the frontal plane motion is given. The paper focuses on the question of push recovery, considering two different cases: when the exoskeleton is pushed as a result of an interaction with another moving object and the case when the exoskeleton stands on a platform that rapidly changes its speed. A push recovery algorithm is proposed that allows the exoskeleton to regain vertical balance by taking one step. The algorithm was tested via numerical simulation; the results are shown and analysed in the paper. The results of the simulation demonstrated the similarity of the exoskeleton motion to that of a human.

An algorithm for generating convex obstacle-free regions based on stereographic projection

Many of the existing motion planning algorithms require the free space to be decomposed into convex obstacle-free regions. This paper presents an algorithm for generating such regions. The algorithm is based on stereographic projection. The algorithm requires relatively simple computations, compared with some of the known methods. Three applications of the algorithm were demonstrated: motion planning for a walking robot, trajectory generation for an unmanned aerial vehicle and a motion planning for an in-pipe robot.

Study on a Two-Staged Control of a Lower-Limb Exoskeleton Performing Standing-Up Motion from a Chair

The paper is concerned with control over a lower-limb exoskeleton device when it performs a sit-to-stand motion from a chair. A mathematical model describing full dynamics of the device is derived, and the strategies to facilitate the desired motion are outlined. A control system based on a modified Jacobian transpose is proposed, and its performance is evaluated in a series of numerical experiments. The simulations proved that the proposed control method can be successfully used to provide stable sit-to-stand motion from a chair.

A Control Strategy for a Lower Limb Exoskeleton with a Toe Joint

In this paper a lower limb exoskeleton with a toe joint is studied. A mathematical model of the exoskeleton is presented, and the equations of motion are given. The exoskeleton is controlled with a feedback controller. The control system attempts to move the center of mass of the exoskeleton along the desired trajectory. To find the joint space trajectory that allows to perform the desired motion a numerical optimization-based iterative algorithm for solving inverse kinematics is given. The algorithm allows to engage and disengage the toe joint, based on how close the mechanism is to a singular position. That gives us an automatic human-like toe joint engagement, that can be controlled though certain parameters, which is discussed in the paper. The results of the numerical simulation of the exoskeleton motion are presented.