|
Costmap Generation Based on Dynamic Obstacle Detection and Velocity Obstacle Layer for Autonomous Mobile Robot |
|
Si Yu Lin, Chin Sheng Chen(National Taipei University of Technology, Taiwan) |
| The environmental conditions corresponding to dangerous or collided areas are generally represented by Costmap when the Autonomous Mobile Robot (AMR) is navigated. Here, this paper provides a Costmap 2D layer plug-in, Velocity Obstacle layer, it can accurately detect obstacle’s coordination and radius and then estimate the obstacle’s velocity to create Velocity Obstacle which can represent the potential collision vector in the future. In the simulation, we assume the robot’s max velocity is 0.2m/s and an obstacle move forward to the robot with 0.3m/s. The results show the AMR can avoid the obstacle well. In experiment, the AMR also can avoid the people moving toward it in the real world |
|
|
Real Time Fault Detection for Mecanum Wheel Omnidirectional Robot Platform |
|
Peter I-Tsyuen Chang(National Taiwan University of Science and Technology, Taiwan) |
| This research demonstrates fault detection of omnidirectional vehicles with Mecanum wheels, when rotational error occurs in operation and in real time, by directly monitoring the viscous frictional coefficient through the torque model of each of the four operational Mecanum wheels. The relationship between the viscous friction coefficients and the speed of each wheels are obtained by both the kinematics and dynamics modeling of the omni-directional vehicle system. Experiments are done for verification, and shows over 5 times change in friction for in-fault situations. |
|
|
Wood Polish Classification for Automated Quality Inspection based on AI Vision |
|
Hsien-I Lin, Satrio Dwi Sanjaya(National Taipei University of Technology, Taiwan) |
| Nowadays, the demand for quality inspection of wood polishing is increasing. Thus, there is a need on industrial level to maintain high quality inspection. The quality inspection on wood polishing is currently done by human labors, which is inefficient, costly, and time-consuming. To reduce the cost of wood quality inspection, we propose an automated quality inspection based on AI vision to distinguish whether the wood is polished or unpolished. This system uses a deep learning method to classify polished or unpolished wood. The result showed the competitive performance metric of 85% as recall, 85.5% as precision, and 85% as f1-score. |
|
|
Interactive Motion Planning for Autonomous Robotic Photo Taking |
|
Kai-Tai Song, Pei-Chun Lu(National Yang Ming Chiao Tung University, Taiwan) |
| In this paper, a motion planning method is proposed for an autonomous photo-taking robot. The developed human-robot interactive motion planning and control system is based on face recognition and facial pose estimation of a target person. The target person’s location is estimated after combined with robot self-localization result. Then the goal position and orientation of the robot are determined for autonomous photo-taking of the specific person. An APP for mobile phones has been integrated to select the target person and transmit the command to the robot for photo taking. After receiving the command, the robot autonomously navigates to the goal location and shoots pictures of the target p |
|
|
Task-Oriented Navigation System for Dynamical Multiple Social Tasks of a Service Robot |
|
Shao-Hung Chan, Siquan Zeng, Li-Chen Fu(National Taiwan University, Taiwan) |
| In this paper, we propose a novel task-oriented navigation system for robots to achieve social interaction tasks with the help of perceptions. To organize these tasks, we propose an instruction structure consisting of decaying rewards concerning priorities and time. Moreover, we model the indoor scenario into a graph structure to allocate instructions and propose a task planning algorithm. With our system, the social robot is able to meet human requirements and further efficiently interact with people in a multiple-human environment, achieving sophisticated human-robot interaction (HRI). |
| |