In the navigation of mobile robots, the driving risk can be minimized by increasing the probability of success. The algorithm, which is currently commonly known as the shortest path algorithm, performs efficiently, but does not exhibit a good probability of success for achieving the final goal. In this paper, we develop a new reactive navigation algorithm, known as the goal guidance vector (G2V), which can minimize the driving risk within the sensing range. The G2V is designed to improve the performance of the reactive navigation algorithm using a hazard cost function (HCF) that accounts for the scale and locations of the obstacles within the sensing range. We also adopt real-time fuzzy reactive control to determine the weighting factors of the HCF in an unknown environment to determine the optimal G2V. Simulations are conducted to validate the use of this approach for various environments.

In this paper, we propose a new driving mode control algorithm for a mobile robot based on obstacle detection. The robot has a variable geometry single-tracked mechanism, so it can maximize a contact length with ground for the adaptability to off-road and puesue a stable system due to the lower center of gravity. However this robot system embodied passive type according to operator. In this reason, several problems are detected. So, this research presents a new method of obstacle detection using PSD infrared sensors and translates the variable tracks on the best suited driving mode actively. And experimental results about mentioned are presented.