목적: 이 연구에서는 장애물 높이에 따른 만성 뇌졸중 환자의 장애물 보행 특성과 보상전략을 파악하였다. 방법: 이를 위해 두 가지 장애물 높이 조건(5cm와 15cm)에 따라 장애물 보행의 운동학적 특성을 정상인과 비교하였다. 구체적으로 장애물 보행 전 이륙 거리, 장애물 통과 시 장애물 통과 높이와 장애물 통과 후 착지 거리를 측정하 였다. 결과: 분석결과, 만성 뇌졸중 환자들은 정상인보다 보행 전 이륙 거리와 장애물 통과 후 착지 거리가 상대 적으로 짧은 것으로 나타났다. 또한, 뇌졸중 환자들은 정상인보다 장애물을 통과할 때 통과 높이가 높은 것으로 밝혀졌다. 이러한 보행 움직임 패턴은 두 가지 장애물 조건에서 모두 유사하게 나타났다. 결론: 이러한 결과는 뇌 졸중 환자들이 편측 마비로 인한 움직임 제약에 기인하는 것으로 판단된다. 또한, 신체적 장애로 인해 뇌졸중 환 자들은 장애물 보행 시 낙상을 예방하고, 안정적으로 장애물 보행을 위한 운동보상전략에서 정상인과 차이를 보 인 것으로 생각된다. 결론적으로 이 연구는 만성 뇌졸중 환자들의 운동보상 전략을 장애물 보행을 통해서 규명했 다는 점에서 현장에 유용한 시사점을 제공할 것으로 기대된다.

As the development of autonomous vehicles becomes realistic, many automobile manufacturers and components producers aim to develop ‘completely autonomous driving’. ADAS (Advanced Driver Assistance Systems) which has been applied in automobile recently, supports the driver in controlling lane maintenance, speed and direction in a single lane based on limited road environment. Although technologies of obstacles avoidance on the obstacle environment have been developed, they concentrates on simple obstacle avoidances, not considering the control of the actual vehicle in the real situation which makes drivers feel unsafe from the sudden change of the wheel and the speed of the vehicle. In order to develop the ‘completely autonomous driving’ automobile which perceives the surrounding environment by itself and operates, ability of the vehicle should be enhanced in a way human driver does. In this sense, this paper intends to establish a strategy with which autonomous vehicles behave human-friendly based on vehicle dynamics through the reinforcement learning that is based on Q-learning, a type of machine learning. The obstacle avoidance reinforcement learning proceeded in 5 simulations. The reward rule has been set in the experiment so that the car can learn by itself with recurring events, allowing the experiment to have the similar environment to the one when humans drive. Driving Simulator has been used to verify results of the reinforcement learning. The ultimate goal of this study is to enable autonomous vehicles avoid obstacles in a human-friendly way when obstacles appear in their sight, using controlling methods that have previously been learned in various conditions through the reinforcement learning.

The Expanded Guide Circle (EGC) method has been originally proposed as the guidance navigation method for improving the efficiency of the remote operation using the sensory information. The previous algorithm is, however, concerned only for the omni-directional mobile robot, so it needs to suggest a suitable one for a mobile robot with non-holonomic constraints. The ego-kinematic transform is a method to map points of R2 into the ego-kinematic space which implicitly represents non-holonomic constraints for admissible paths. Thus, robots with non-holonomic constraints in the ego-kinematic space can be considered as “free-flying object”. In this paper, we propose an effective obstacle avoidance method for mobile robots with non-holonomic constraints by applying EGC method in the ego-kinematic space using the ego-kinematic transformation. This proposed method shows that it works better for non-holonomic mobile robots such as differential-drive robot than the original one. The simulation results show its effectiveness of performance.

Many intelligent robots have to be given environmental information to perform tasks. In this paper an intelligent robot, that is, a cleaning robot used a sensor fusing method of two sensors: LRF and StarGazer, and then was able to obtain the information. Throughout wall following using laser displacement sensor, LRF, the working area is built during the robot turn one cycle around the area. After the process of wall following, a path planning which is able to execute the work effectively is established using flow network algorithm. This paper describes an algorithm for minimal turning complete coverage path planning for intelligent robots. This algorithm divides the whole working area by cellular decomposition, and then provides the path planning among the cells employing flow networks. It also provides specific path planning inside each cell guaranteeing the minimal turning of the robots. The proposed algorithm is applied to two different working areas, and verified that it is an optimal path planning method.

This paper presents a goal-directed reactive obstacle avoidance method based on lane method. The reactive collision avoidance is necessarily required for a robot to navigate autonomously in dynamic environments. Many methods are suggested to implement this concept and one of them is the lane method. The lane method divides the environment into lanes and then chooses the best lane to follow. The proposed method does not use the discrete lane but chooses a line closest to the original target line without collision when an obstacle is detected, thus it has a merit in the aspect of running time and it is more proper for narrow corridor environment. If an obstacle disturbs the movement of a robot by blocking a target path, a robot generates a temporary target line, which is parallel to an original target line and tangential to an obstacle circle, to avoid a collision with an obstacle and changes to and follows that line until an obstacle is removed. After an obstacle is clear, a robot returns to an original target line and proceeds to the goal point. Obstacle is recognized by laser range finder sensor and represented by a circle. Our method has been implemented and tested in a corridor environment and experimental results show that our method can work reliably.

In this paper, we provide experimental results and verification for obstacle avoidance algorithm 'ELA(Emergency Level Around)', which is applicable to rescue robots. ELA is a low level intelligence-based obstacle avoidance algorithm, so can be used in fast mobile robots requiring high speed in operation with little computational load. Constructed system for experiments consist of laptop, sensors, peripheral devices and mobile robot platform VSTR(Variable Single-tracked Robot) to realize predetermined scenarios. Finally, experiment was conducted in indoor surroundings including miscellaneous things as well as dark environment to show fitness and robustness of ELA for rescue, and it is shown that VSTR navigates endowed area well with real-time obstacle avoidance based on ELA. Therefore, it is concluded that ELA can be a candidate algorithm to increase mobility of rescue robots in real situation.

Collision avoidance is a fundamental and important task of an autonomous mobile robot for safe navigation in real environments with high uncertainty. Obstacles are classified into static and dynamic obstacles. It is difficult to avoid dynamic obstacles because the positions of dynamic obstacles are likely to change at any time. This paper proposes a scheme for vision-based avoidance of dynamic obstacles. This approach extracts object candidates that can be considered moving objects based on the labeling algorithm using depth information. Then it detects moving objects among object candidates using motion vectors. In case the motion vectors are not extracted, it can still detect the moving objects stably through their color information. A robot avoids the dynamic obstacle using the dynamic window approach (DWA) with the object path estimated from the information of the detected obstacles. The DWA is a well known technique for reactive collision avoidance. This paper also proposes an algorithm which autonomously registers the obstacle color. Therefore, a robot can navigate more safely and efficiently with the proposed scheme.

We propose a novel real-time obstacle avoidance method for rescue robots. This method, named the ELA(Emergency Level Around), permits the detection of unknown obstacles and avoids collisions while simultaneously steering the mobile robot toward safe position. In the ELA, we consider two sensor modules, PSD(Position Sensitive Detector) infrared sensors taking charge of obstacle detection in short distance and LMS(Laser Measurement System) in long distance respectively. Hence if a robot recognizes an obstacle ahead by PSD infrared sensors first, and judges impossibility to overcome the obstacle based on driving mode decision process, the order of priority is transferred to LMS which collects data of radial distance centered on the robot to avoid the confronted obstacle. After gathering radial information, the ELA algorithm estimates emergency level around a robot and generates a polar histogram based on the emergency level to judge where the optimal free space is. Finally, steering angle is determined to guarantee rotation to randomly direction as well as robot width for safe avoidance. Simulation results from wandering in closed local area which includes various obstacles and different conditions demonstrate the power of the ELA.