PATH PLANNING AND FORMATION CONTROL VIA POTENTIAL FUNCTION FOR UAV QUADROTOR
AHMAD ATAKA AWWALUR RIZQI, Dr. Eng. Adha Cahyadi, ST., M.Eng.
2014 | Skripsi | TEKNIK ELEKTROPada penelitian ini dirancang sistem kendali untuk path planning dan kendali formasi pada Quadrotor berbasis potential function. Potential function digunakan untuk mengatur gerakan Quadrotor menuju tempat yang diinginkan serta menghindari rintangan. Algoritma untuk menyelesaikan masalah local minima dalam path planning dengan menggunakan strategi wall-following juga dirancang. Hasil kendali path planning ini kemudian digunakan sebagai dasar perancangan kendali formasi. Menggunakan skema gabungan antara virtual leader dan pendekatan behavioral, kendali formasi berbasis potential function juga dibahas. Setiap Quadrotor diberi attractive potential untuk mencapai posisi tertentu dalam formasi. Setiap Quadrotor juga dianggap sebagai rintangan bergerak yang menghasilkan repulsive potential untuk menghindari tabrakan antar robot. Posisi Quadrotor pada setiap konfigurasi formasi dihitung berdasarkan posisi virtual leader. Virtual leader menggunakan model robot unicycle sebagai persamaan geraknya. Repulsive potential juga diberikan pada setiap agen ketika memasuki wilayah terjangkau dari rintangan. Seluruh strategi kendali yang diajukan berhasil diaplikasikan pada model Quadrotor Parrot AR Drone 2.0 di Simulator Gazebo yang terprogram lewat Robot Operating System.
Potential-field-based control strategy for path planning and formation control of multi Quadrotor systems is proposed in this work. The potential field is used to attract the Quadrotor to the goal location as well as avoiding the obstacle. The algorithm to solve the so called local minima problem by utilizing the wall-following behavior is also proposed. The resulted path planning via potential function strategy is then used to design formation control algorithm. Using the virtual leader approach, the formation control strategy by means of potential function is formulated. Each Quadrotor is assigned attractive potential to reach its desired position in formation. Each Quadrotor is also treated as moving obstacle which induces repulsive potential to the others in order to prevent inter-robot collision. The desired position for each Quadrotor in certain configuration is calculated based on the virtual leader’s current position. To get smooth movement, the virtual leader is set as unicycle robot. The repulsive potential is also attached to each agent in case it moves in the region inside the obstacle’s influence. The overall strategy has been successfully applied to the Quadrotor’s model of Parrot AR Drone 2.0 in Gazebo simulator programmed using Robot Operating System.
Kata Kunci : Path Planning, Kendali Formasi, Potential Function, Local Minima, Quadrotor
