No abstract

No abstract

This paper describes a trajectory tracking method using environmental magnetic field for outdoor autonomous mobile robots. In this research, a 3-axis magnetic sensor is used to scan DC magnetic field in the outdoor environment to build a database. The robot then performs trajectory tracking based on the database. The experimental results show that by applying the proposed method the robot is possibly able to navigate in the outdoor environment with a reliable accuracy. S. A. Rahok is Ph. D. Candidate with

No abstract

In mobile robot's localization, it is well known that odometry can provide a reliable accuracy in short term navigation and a very high sampling rate. However, odometry produces cumulative error because of uneven terrains or wheel slippage and this error increases proportionally with the distance traveled by the mobile robot. Therefore, it is necessary to augment odometry with other sensors to improve its accuracy. This paper proposes an estimation method of mobile robot orientation using an environmental magnetic field (magnetic field that occurs in the environment). A three-axis magnetic sensor is utilized to scan the environmental magnetic field to build a magnetic database on a grid map called "a magnetic map" with the mobile robot operated with a joystick on a desired route. The mobile robot then estimates its orientation by comparing the magnetic sensor readings with the magnetic data stored in the magnetic map. However, even if the proposed method can improve the accuracy of the odometry, positioning error still remains as a major problem in long term navigation. In this work, a localization method using Monte Carlo Localization (MCL) based on a Light Detection and Ranging (LIDAR) is utilized to fix the positioning error at the areas where landmark can be observed. The experimental results showed that the mobile robot could localize robustly in any environments with the proposed method.

No abstract

Functionality is not the only major objective of service robots design, but safety and affinity are also necessary. This study focuses on a design policy for a service robot that can be used to transport people in a real world environment. In this paper, a design policy which referred to the rules of a real world robot challenge (RWRC) called "Tsukuba Challenge 2010" and an idea to design the robot's body which is suitable for navigation method based on internal sensors, are presented. The design policy emphasizes functionality, safety, and affinity. Therefore, a service robot which can ensure the functionality and the safety during coexistence with humans can be developed. In 2010, we developed a robot with affinity and applied it in many events including Tsukuba Challenge. A questionnaire on the robot's perfection level was taken during the events. Feasibility of the proposed policy is shown by the results and performance of the demonstrations.

No abstract