Akademik Veri Yönetim Sistemi
26th IEEE Signal Processing and Communications Applications Conference, SIU 2018, İzmir, Türkiye, 2 - 05 Mayıs 2018, ss.1-4
In this study, the subject of moving obstacle detection which is frequently studied in mobile robot systems is examined. Objects at speeds and sizes close to real-world conditions have been identified and classified using a lidar sensor model. The next position was estimated from the previous and the current position information of each obstacle. The trajectory planner decided not according to the instant obstacle profile but according to the predicted obstacle configuration. Thus, the moving obstacle is prevented from misdirecting the mobile robot. The working environment is a toolbox previously developed by the authors for Matlab. A model of the LMS100 series lidar sensor and a mobile robot with a differential drive are also defined in this Mobile Robot Module. Similarly, the gap detection and the optimal local target planning modules are also defined as embedded in the Toolbox.