The convolutional neural network (CNN) is a technique that often used in deep learning. Various models have been proposed and improved for learning on CNN. When with CNN, it important to determine the optimal parameters. This paper proposes an optimization of CNN parameters using logarithm decreasing inertia weight (LogDIW). two datasets, i.e., MNIST CIFAR-10 dataset. experiment, dataset, compa...

Balance beam, a controller that is built adopting the gyro effect, can freely control the attitude of an unstructured object by means of the position controlling of an inner gimbal. However in the former research, the weight (inertia) of a load should be known to issue a proper command for the gimbal, which makes it hard to operate the balance beam in field applications because the load can be ...

Considering the issues of premature convergence and low solution accuracy in solving high-dimensional problems with basic chicken swarm optimization algorithm, an adaptive simplified algorithm based on inverted S-shaped inertia weight (ASCSO-S) is proposed. Firstly, a presented by removing all chicks from swarm. Secondly, designed introduced into updating process roosters hens to dynamically ad...

Gait rehabilitation using body weight support on a treadmill is a recommended rehabilitation technique for neurological injuries, such as spinal cord injury. In this paper, a new robotic orthosis is presented for treadmill training. In the presented design the criteria such as low inertia of robot components, backdrivability, high safety and degrees of freedom based on human walking are conside...