Visual Flight Control of a Quadrotor Using Bioinspired Motion Detector

Abstract

Motion detection in the fly is extremely fast with low computational requirements. Inspired from the fly's vision system, we focus on a real-time flight control on a miniquadrotor with fast visual feedback. In this work, an elaborated elementary motion detector (EMD) is utilized to detect local optical flow. Combined with novel receptive field templates, the yaw rate of the quadrotor is estimated through a lookup table established with this bioinspired visual sensor. A closed-loop control system with the feedback of yaw rate estimated by EMD is designed. With the motion of the other degrees of freedom stabilized by a camera tracking system, the yaw-rate of the quadrotor during hovering is controlled based on EMD feedback under real-world scenario. The control performance of the proposed approach is compared with that of conventional approach. The experimental results demonstrate the effectiveness of utilizing EMD for quadrotor control.

Fly visual system inspired artificial neural network for collision detection

“…Motion detection on direction and velocity has been being an active research topic in visual neuroscience since 1956 [15][16][17][18][19][20][21][22][23], in which estimation on position and velocity is a basic and vital branch. Nakamura [15] and Ruffier et al [16] studied optic flow and direction estimators.…”

“…Visual perception and motion detection are essential and crucial in processing visual information or optic flow that guides much of fly's behavior [6,7]. Studies on such visual system can be categorized into at least four broader types: internal neuronal structure anatomy [1][2][3][4][5]; computational models [8][9][10][11][12][13][14]; motion detection and tracking [15][16][17][18][19][20][21][22][23]; and motion control [19,20]. Especially, visual neuroscientists, who reported many valuable neuronal-based circuit models [1][2][3][4][5], made great efforts to explore intensively internal relationships between neurons, e.g., large monopolar cell (LMC), m 1 and m 2 cells, and lobula plate tangential cell (LPTC).…”

“…Especially, visual neuroscientists, who reported many valuable neuronal-based circuit models [1][2][3][4][5], made great efforts to explore intensively internal relationships between neurons, e.g., large monopolar cell (LMC), m 1 and m 2 cells, and lobula plate tangential cell (LPTC). These provide researchers with extremely useful bio-inspirations in resolving complex engineering problems, e.g., object tracking [21][22][23], flight control [19,20], and colliding avoidance and detection [24][25][26][27][28][29].…”

“…Can it be utilized to construct artificial visual neural networks for one such problem? Two recent achievements [19,20] have indicated that EMD is a potential tool for detecting the change of optic flow arisen from an imminent object. Particularly, Misserler and Kamangar [23] proposed an artificial neural network (ANN) for detecting and tracking a moving object within the field of view, instead of collision detection.…”

A Collision Control Strategy for Multiple Moving Robots

“…The sensor is located in front (i.e., the moving direction). Vision is used in many applications such as optical motion planning mapping [6], [8], or the motion sensors themselves [4], [3], [11].…”