Search

The toll of human casualties and psychological impacts on societies make any study on violent extremism worthwhile, let alone attempting to detect patterns among them. This paper is an effort to predict which violent extremist organization (VEO), among 14 currently active ones throughout the world, is responsible for a violent act based on 14 features, including its human and structural tolls, its target type and value, intelligence, and weapons utilized in the attack. Three main steps in our paper include: 1) the visualization of the violent acts through linear and non-linear dimensionality reduction techniques; 2) sequential forward feature selection based on the generalization accuracy of three machine learning models-decision tree, and linear and nonlinear SVM; and 3) employing multilayer perceptron to predict the VEO based on the selected features of a violent act. Top-ranked selected features were related to the target type and plan and the multilayer perceptron achieved up to 40% test accuracy.

The lower limbs of the human body actually can perform the multiple-degree-of-freedom motion, just like the upper limbs. This suggests the possibility for the lower limbs to be used in the operation of multiple-degree-of-freedom devices, such as a robot arm. With that point in mind, the present paper focuses on the foot motion and examines its characteristics under the situation in which the posture of the object is manipulated by the posture change of the foot. First, we investigated how well the foot of the operator moved in accordance with the intention of the operator in order to clarify the motion characteristics of the foot experimentally by measuring the foot motion with a motion capture system under the assumption that the operator manipulates an object in virtual space. The results showed that there are differences between the intended and actual foot motions, especially when the tilt angle change was accompanied by a rotation angle change, which might be because of the joints whose axes of motion are nonparallel to the foot coordinate system, such as the talocalcaneal joint or Chopart joint. Next, an operating system considering the motion characteristics of the foot was proposed, and an experiment to verify its effectiveness was conducted. When the proposed conversion formula was used to calculate the intended foot motion based on the actual foot motion, the operability improved with respect to the required time and path-following accuracy while manipulating an object to the target posture and with respect to subjective operability.

Robots are meant to replace humans for a broad variety of everyday tasks, such as environmental monitoring or patrolling large public areas for security assurance. The main focus of researchers and practitioners has been on providing tailored software and hardware solutions for very specific and often complex tasks. On one hand, these solutions show great potential and provide advanced capabilities for solving the specific task. On the other hand, the polarized attention to task-specific solutions makes them hard to reuse, customize, and combine. In this paper we propose a family of domain-specific modeling languages for the specification of civilian missions of mobile multi-robot systems. These missions are meant to be described in terms of models that are: 1) closer to the general problem domain; 2) independent from the underlying technologies; 3) ready to be analyzed, simulated, and executed; and 4) extensible to new application domains, thus opening up the use of robots to even non-technical operators. Moreover, we show the applicability of the proposed family of languages in two real-world application domains: unmanned multicopters and autonomous underwater vehicles.