• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.168-173
• /
• 2003

In this paper, we will present a deeper look on optimal design methods that are related to path-planning for a mobile robot. To control the motion of a mobile robot in a clustered environment, it's necessary to know a suitable trajectory assuming certain start and goal point. Up to now, there are many literatures that concern optimal path planning for an obstacle avoided mobile robot. Among those literatures, we have chosen 2 novel methods for our further analysis. The first approach [4] is based on HJB(Hamilton-Jacobi-Bellman) equation whose solution is the return-function that helps to generate a shortest path to the goal. The later [5] is called polynomial-path-planning approach, in this method, a shortest polynomial-shape path would become a solution if it was a collision-free path. The camera network plays the role as sensors to generate updated map which locates the static and dynamic objects in the space. Therefore, the exhibition of both path planning and dynamic obstacle avoidance by the updated map would be accomplished simultaneously. As we mentioned before, our research will include the motion control of a true mobile robot on those optimal planned paths which were generated by above algorithms. Base on the kinematic and dynamic simulation results, we can realize the affection of moving speed to the stable of motion on each generated path. Also, we can verify the time-optimal trajectory through velocity tuning. To simplify for our analysis, we assumed the obstacles are cylindrical circular objects with the same size.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.13 no.3
• /
• pp.225-231
• /
• 2003

A new three-dimensional thin shell element for a structure containing an integrated distributed piezoelectric sensor and actuator is Proposed. The assumed strain formulation and the bubble function are introduced to improve the performance of the shell element. A finite element formulation gives a general tool that can predict the static or dynamic responses of the shell with piezoelectric sensor/actuator. The verification through the calculation of the static response for the piezoelectric bimorph beam shows that the results agree with those from the theoretical analysis very well. Dynamic response of a shell shows that the reduction of vibration is possible with the introduction of the piezoelectric shell sensor and actuator. However. the curvature of sensor/actuator is an obstacle for application, since the flexible PVDF is not strong enough and the PZT with curvature should be made specially.

• Journal of Institute of Control, Robotics and Systems
• /
• v.16 no.3
• /
• pp.233-238
• /
• 2010

This paper describes a multi-mobile robot system with fuzzy rule based structure in collision avoidance. Collision avoidance is an important function to perform a given task collaboratively and cooperatively in multi-mobile robot environments. So the important but challenging problem is handled in this paper. Considered obstacles for collision avoidance between multi mobile robots are static, dynamic, or both of them at the same time. Using the fuzzy rule based structure, distance and angle from a robot to obstacles are described as fuzzy linguistic values and steering angle for the robot are updated from the collision environments. As a result, the multi-mobile robot can modify a global path from a robot itself to its own target. In addition, avoiding collision with static or dynamic obstacles for the robot system can be achieved. Simulation based experimental results are given to show usefulness of this method.

• Advances in robotics research
• /
• v.1 no.1
• /
• pp.21-39
• /
• 2014

Due to the rapid progress in the field of robotics, it is a high time to concentrate on the development of a robot that can manoeuvre in all type of landscapes, ascend and descend stairs and sloping surfaces autonomously. This paper presents details of a prototype robot which can navigate in very rough terrain, ascend and descend staircase as well as sloping surface and cross ditches. The robot is made up of six differentially steered wheels and some passive mechanism, making it suitable to cross long ditches and landscape undulation. Static stability of the developed robot have been carried out analytically and navigation capability of the robot is observed through simulation in different environment, separately. Description of embedded system of the robot has also been presented and experimental validation has been made along with some details on obstacle avoidance. Finally the limitations of the robot have been explored with their possible reasons.

• Journal of the Korean Society for Precision Engineering
• /
• v.13 no.11
• /
• pp.152-162
• /
• 1996

A four-legged Walking Machine can move on the plain terrain with mobility and stability and stability, but if there exist any obstacles on the terrain of the motion direction, it takes extra time to cross those obstacles and the stability should be considered during motion. The main objective is the study a Quadruped which can cross obstacles with better mobility, stability and fuel economy than any other wheeled or tracked vehicles. Vertical step, isolated wall and ditch are the basic obstacles and by understanding those three cases perfectly, a Quadruped can move on any mixed rough terrain as 4-legged terrestrial vertebrates move. Each leg of a Quadruped has a limited walk space called a walking volume and this is very important to deter- mine the crossing capability in a static analysis. A Quadruped can be simplified with links and joints. By applying the research method, a quadruped can determine the control procedures as soon as it receives the terrain information from scanner and finally can move with mobility and stability.

• ETRI Journal
• /
• v.37 no.5
• /
• pp.1001-1011
• /
• 2015

Since just-in-time (JIT) has considerable overhead to detect hot spots and compile them at runtime, using sophisticated optimization techniques for embedded devices means that any resulting performance improvements will be limited. In this paper, we introduce a novel static Dalvik bytecode optimization framework, as a complementary compilation of the Dalvik virtual machine, to improve the performance of Android applications. Our system generates optimized Dalvik bytecodes by using Low Level Virtual Machine (LLVM). A major obstacle in using LLVM for optimizing Dalvik bytecodes is determining how to handle the high-level language features of the Dalvik bytecode in LLVM IR and how to optimize LLVM IR conforming to the language information of the Dalvik bytecode. To this end, we annotate the high-level language features of Dalvik bytecode to LLVM IR and successfully optimize Dalvik bytecodes through instruction selection processes. Our experimental results show that our system with JIT improves the performance of Android applications by up to 6.08 times, and surpasses JIT by up to 4.34 times.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.3
• /
• pp.15-20
• /
• 2024

Path planning for unmanned aerial vehicles (UAV) is crucial in avoiding collisions with obstacles in complex environments that include both static and dynamic obstacles. Path planning algorithms like RRT and A^* are effectively handle static obstacle avoidance but have limitations with increasing computational complexity in high-dimensional environments. Reinforcement learning-based algorithms can accommodate complex environments, but like traditional path planning algorithms, they struggle with training complexity and convergence in higher-dimensional environment. In this paper, we proposed a reinforcement learning model utilizing a cell decomposition algorithm. The proposed model reduces the complexity of the environment by decomposing the learning environment in detail, and improves the obstacle avoidance performance by establishing the valid action of the agent. This solves the exploration problem of reinforcement learning and improves the convergence of learning. Simulation results show that the proposed model improves learning speed and efficient path planning compared to reinforcement learning models in general environments.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.2
• /
• pp.163-173
• /
• 2013

We present a new computer vision technique which can detect unexpected or static vertical objects in road driving environment. We first obtain temporal and spatial difference images in each frame of a stereo video sequence. Using the difference images, we then generate VO and HO maps by improving the conventional V and H disparity maps. From the VO and HO maps, candidate areas of vertical obstacles on the road are detected. Finally, the candidate areas are merged and refined to detect vertical obstacles.

• Journal of Auto-vehicle Safety Association
• /
• v.11 no.2
• /
• pp.17-22
• /
• 2019

This paper describes the tracking algorithm performance evaluation for autonomous vehicle using a safety envelope based path. As the level of autonomous vehicle technologies evolves along with the development of relevant supporting modules including sensors, more advanced methodologies for path generation and tracking are needed. A safety envelope zone, designated as the obstacle free regions between the roadway edges, would be introduced and refined for further application with more detailed specifications. In this paper, the performance of the path tracking algorithm based on the generated path would be evaluated under safety envelop environment. In this process, static obstacle map for safety envelope was created using Lidar based vehicle information such as current vehicle location, speed and yaw rate that were collected under various driving setups at Seoul National University roadways. A level of safety was evaluated through CarSim simulation based on paths generated with two different references: a safety envelope based path and a GPS data based one. A better performance was observed for tracking with the safety envelop based path than that with the GPS based one.

• Journal of Drive and Control
• /
• v.19 no.4
• /
• pp.110-116
• /
• 2022

Recently, various studies have been conducted to apply deep learning and AI to various fields of autonomous driving, such as recognition, sensor processing, decision-making, and control. This paper proposes a controller applicable to path following, static obstacle avoidance, and pedestrian avoidance situations by utilizing reinforcement learning in autonomous vehicles. For repetitive driving simulation, a reinforcement learning environment was constructed using virtual environments. After learning path following scenarios, we compared control performance with Pure-Pursuit controllers and Stanley controllers, which are widely used due to their good performance and simplicity. Based on the test case of the KNCAP test and assessment protocol, autonomous emergency steering scenarios and autonomous emergency braking scenarios were created and used for learning. Experimental results from zero collisions demonstrated that the reinforcement learning controller was successful in the stationary obstacle avoidance scenario and pedestrian collision scenario under a given condition.