|
Active Sampling for Maximum seeking and environment field reconstruction using cooperative mobile sensors |
|
Jaemin Seo(UNIST, Korea), Geunsik Bae, Yeongho Song, Hyondong Oh(Ulsan National Institute of Science and Technology, Korea) |
| This paper presents an active sampling algorithm that finds the maximum point while reconstructing the unknown environment field using cooperative mobile sensors. Bayesian optimization is employed as an active sampling, and Gaussian processes is used for environment field reconstruction. The mobile sensors move in a direction that maximizes the mutual information to the maximum point for active data collection. For an efficient cooperation strategy of mobile sensors, a potential game approach with joint strategy fictitous play is adopted for negotiating joint actions among mobile sensors. |
|
|
Frequency-domain Analysis of Precision Altitude Controller for a Fixed-wing UAV using Virtual Target Algorithm |
|
Sejin Jang, Jaehun Jeong, Yeonghwan Jang, Hoijo Jeaong, Seungkeun Kim, Jinyoung Suk(Chungnam National University, Korea) |
| In this paper, a virtual target algorithm is applied to an altitude control of a fixed-wing unmanned aerial vehicle. A transfer function of the altitude output for the command is derived in the discrete-time domain. Using the bode plot, the phase and magnitude are analyzed according to the vehicle ground speed, controller delay, and distance to a virtual target in the frequency domain. To verify the derived transfer function, three degree of freedom simulation and processor in the loop simulation are performed for comparative study. |
|
|
Design of a Drone-Hawk Gripper for Nest Removal from Electric Tower |
|
Bhivraj Suthar(Chungnam Nationl University, Korea), Seul Jung(Chungnam National University, Korea) |
| This paper deals with the bird's nest cleaning equipment, particularly a new type of bird's nest-removing Hawk claw gripper for power transmission towers. Bird's nests are often built on the transmission and distribution network towers. The wooden branches, grass stems, wires and other materials that birds use to make the nest can easily cause the short connection of the power line, causing fire or even burning out the equipment. Causing a power line failure makes a great impact on the economy and human lives. This paper proposed a Hawk-inspired bird's nest cleaning gripper design for a drone arm with three fingers with a long curved talon. The gripper finger talon can penetrate deeper into the nest structure to grip firmly and a drone moves it. The hawk gripper can be easily connected to a foldable drone arm and operated remotely on the electrical tower. We developed a prototype of a hawk gripper and demonstrated its feasibility. |
|
|
Altitude Tracking of UAV with Pitch-Hold Constraint Based on Model Predictive Control for Mine Detection |
|
Hyeong-gwan Kang, Sangmin Lee, Jihoon Lee, Youdan Kim(Seoul National University, Korea), Jinyoung Suk, Seungkeun Kim(Chungnam National University, Korea) |
| A longitudinal altitude and attitude controller is designed for an unmanned aerial vehicle to detect landmines. The UAV considered in this study has two front rotors, two rear rotors, and elevons. Due to the characteristics of the mine detection sensor, the allowable pitch change of the UAV is strictly limited. Therefore, the attitude controller considering the pitch-hold constraint should be designed. To deal with a flight path angle change limit due to the pitch-hold constraint and accurate altitude tracking, a linear model predictive control method is used for altitude control design. Numerical
simulation is carried out to demonstrate the performance of the proposed controller. |
|
|
Effectiveness Analysis on Low-Pass Filter for Inner-loop Attitude Control of Cylindrical Coaxial UAV |
|
Hoijo Jeaong, Taegyun Kim, Joonyoung Hwang, Jaeha Ryi, Seungkeun Kim, Jong-Soo Choi, Jinyoung Suk(Chungnam National University, Korea) |
| This study performs the low-pass filter tuning for inner-loop attitude control of the PX4 autopilot to improve the attitude tracking performance of a coaxial UAV against structure vibration. First, the noise region except for the motor rotation speed region was identified through the fast Fourier transform. Second, according to the low-pass filter's cutoff frequency, the gyro data was visualized as a power spectrogram to analyze the difference in amplitude about the motor rotation speed along the time and frequency axes. Finally, waypoint flight was performed based on the above-tuned values. |
| |