Browsing by Author "Babic, J."
Now showing 1 - 5 of 5
- Results Per Page
- Sort Options
ArticlePublication Open Access Active compliance control reduces upper body effort in exoskeleton-supported walking(IEEE, 2020-04) Uğurlu, Regaip Barkan; Oshima, H.; Sariyildiz, E.; Narildyo, T.; Babic, J.; Mechanical Engineering; UĞURLU, Regaip BarkanThis article presents a locomotion controller for lower limb exoskeletons so as to enable the combined robot and user system to exhibit compliant walking characteristics when interacting with the environment. This is of critical importance to reduce the excessive ground reaction forces during the walking task execution with the aim of improved environmental interaction capabilities. In robot-aided walking support for paraplegics, the user has to actively use his/her upper limbs via crutches to ensure overall balance. By virtue of this requisite, several issues may particularly arise during touchdown instants, e.g., upper body orientation fluctuates, shoulder joints are subject to excessive loading, and arms may need to exert extra forces to counterbalance these effects. In order to reduce the upper body effort via compliant locomotion, the controller is designed to manage the force/position tradeoff by using an admittance controller in each joint. For proof of concept, a series of exoskeleton-aided walking experiments were conducted with the participation of nine healthy volunteers, four of whom additionally walked on an irregular surface for further performance evaluation. The results suggest that the proposed locomotion controller is advantageous over conventional high-gain position tracking in decreasing undesired oscillatory torso motion and total arm force, adequately reducing the required upper body effort.Conference paperPublication Metadata only Assessments on the improved modelling for pneumatic artificial muscle actuators(IEEE, 2015) Peternel, L.; Uğurlu, Regaip Barkan; Babic, J.; Morimoto, J.; Mechanical Engineering; UĞURLU, Regaip BarkanIn this paper, we present an analysis regarding the pneumatic air muscle modelling, with a particular emphasis on the exoskeleton robot control. We propose two calibration approaches for obtaining the model identification data. We used the measurement data acquired from the proposed approaches to identify different mathematical models of pneumatic muscles. These models specified the necessary muscle control pressure for the desired muscle force at a given muscle length value. We compared the performance between the different types of models identified by either of the calibration method. The identified model with the lowest validation error was implemented in pneumatic muscle control for an elbow exoskeleton system. As a result, the system exhibited satisfactory torque and position control tasks, adequately validating the proposed approach.ArticlePublication Metadata only Human motor adaptation in whole body motion(2016) Babic, J.; Öztop, Erhan; Kawato, M.; Computer Science; ÖZTOP, ErhanThe main role of the sensorimotor system of an organism is to increase the survival of the species. Therefore, to understand the adaptation and optimality mechanisms of motor control, it is necessary to study the sensorimotor system in terms of ecological fitness. We designed an experimental paradigm that exposed sensorimotor system to risk of injury. We studied human subjects performing uncon- strained squat-to-stand movements that were systematically subjected to non-trivial perturbation. We found that subjects adapted by actively compensating the perturbations, converging to movements that were different from their normal unperturbed squat-to-stand movements. Furthermore, the adapted movements had clear intrinsic inter-subject differences which could be explained by different adapta- tion strategies employed by the subjects. These results suggest that classical optimality measures of physical energy and task satisfaction should be seen as part of a hierarchical organization of optimality with safety being at the highest level. Therefore, in addition to physical energy and task fulfillment, the risk of injury and other possible costs such as neural computational overhead have to be considered when analyzing human movement.Conference paperPublication Metadata only Inferring cost functions using reward parameter search and policy gradient reinforcement learning(IEEE, 2021) Arditi, Emir; Kunavar, T.; Ugur, E.; Babic, J.; Öztop, Erhan; Computer Science; ÖZTOP, ErhanThis study focuses on inferring cost functions of obtained movement data using reward parameter search and policy gradient based Reinforcement Learning (RL). The behavior data for this task is obtained through a series of squat-to-stand movements of human participants under dynamic perturbations. The key parameter searched in the cost function is the weight of total torque used in performing the squat-to-stand action. An approximate model is used to learn squat-to-stand movements via a policy gradient method, namely Proximal Policy Optimization(PPO). A behavioral similarity metric based on Center of Mass(COM) is used to find the most likely weight parameter. The stochasticity in the training result of PPO is dealt with multiple runs, and as a result, a reasonable and a stable Inverse Reinforcement Learning(IRL) algorithm is obtained in terms of performance. The results indicate that for some participants, the reward function parameters of the experts were inferred successfully.ArticlePublication Metadata only Teaching robots to cooperate with humans in dynamic manipulation tasks based on multi-modal human-in-the-loop approach(Springer Science+Business Media, 2014-01) Peternel, L.; Petric, T.; Öztop, Erhan; Babic, J.; Computer Science; ÖZTOP, ErhanWe propose an approach to efficiently teach robots how to perform dynamic anipulation tasks in cooperation with a human partner. The approach utilises human sensorimotor learning ability where the human tutor controls the robot through a multi-modal interface to make it perform the desired task. During the tutoring, the robot simultaneously learns the action policy of the tutor and through time gains full autonomy. We demonstrate our approach by an experiment where we taught a robot how to perform a wood sawing task with a human partner using a two-person crosscut saw. The challenge of this experiment is that it requires precise coordination of the robot’s motion and complianceaccording to the partner’s actions. To transfer the sawing skill from the tutor to the robot we used Locally Weighted Regression for trajectory generalisation, and adaptive oscillators for adaptation of the robot to the partner’s motion.