This paper presents a ground-based navigation method for an indoor robot to reach a predetermined target. By mining depth maps effectively, a mobile robot could find right path and self-locate with a proposed algorithm named Always Move Straight to the Destination (AMDS). | Journal of Science & Technology 131 (2018) 062-068 An Improved Navigation Method for Robot in Indoor Dynamic Environment Based on Ground Extraction Dang Khanh Hoa, Than Viet Duc, Do Trong Anh, Vu Song Tung, Le Dzung, Nguyen Tien Dzung 1,* Hanoi University of Science and Technology, No. 1, Dai Co Viet, Hai Ba Trung, Hanoi, Viet Nam Received: September 13, 2018; Accepted: November 26, 2018 Abstract This paper presents a ground-based navigation method for an indoor robot to reach a predetermined target. By mining depth maps effectively, a mobile robot could find right path and self-locate with a proposed algorithm named Always Move Straight to the Destination (AMDS). The proposed navigation system extracts the ground plane from the depth map provided by an RGB-D camera. Then the navigation system has established an optimal obstacle avoiding strategy with a success rate of which is better than some recent comparison methods based on Artificial Neural Network (ANN) classifiers or method of combination of two algorithm of Dynamic Window Approach (DWA) and Anytime Repairing A* (ARA*). The robot's navigational capability is more flexible than the comparison methods because the angle of direction adjustment is 1 degree. The proposed ground-based navigation method could be integrated into low cost robots. Keywords: depth map, ground plane, navigation, path direction robots with abilities of detection surroundings, obstacle-avoiding and moving toward the target area. But the robot only can reach the target area rather than target point. And the robot needs a powerful sensor that can collect data likely a camera. 1. Introduction* In recent years, mobile robots are being used in the community widely to support vulnerable individuals such as children or the elderly or the sick persons. These subjects require careful care with the supported system requirements to be aware of the context in which they are operating. The data acquisition automation can only meet the .