|
Semantic Indoor Scene Recognition of Time-Series Arial Images from a Micro Air Vehicle Mounted Monocular Camera |
|
Hirokazu Madokoro, Shinya Ueda, Kazuhito Sato(Akita Prefectural University, Japan) |
| This paper presents a semantic scene recognition method from indoor areal time-series images obtained using a micro air vehicle (MAV). Using category maps, topologies of image features are mapped into a low-dimensional space based on competitive and neighborhood learning. The experimentally obtained results with leave-one-out cross-validation (LOOCV) for datasets divided with 10 zones revealed respective mean recognition accuracies for the round flight datasets and zigzag flight datasets of 71.7% and 65.5%. |
|
|
Path tracking control of AUV using nonholonomic error dynamics |
|
Gun Rae Cho(Korea Institute of Robot and Convergence, Korea) |
| The paper proposes a control scheme for path tracking of Autonomous Underwater Vehicle(AUV), which can handle the issue regarding non-holonomic constraint and that regarding nonlinear dynamics of AUV separately. The non-holonomic issue is handled by designing the desired error dynamics appropriately; whereas, nonlinear dynamics of AUV is compensated by employing a robust feedback controller. Not requiring whole dynamic model of AUV, the proposed scheme has simple structure and is easy to design. Through the simulation results, the tracking performance of the proposed controller is verified. |
|
|
Comparative Study of the Performance of Application of Bio-Inspired Strategies to Pursuit Evasion Game Under Feedback Laws |
|
Lairenjam Obiroy Singh(Hindustan Institute of Technology and Science, India) |
| Pursuit Evasion Game (PEG) is an abstract model of various significant problems that appear in both civil and military applications. Bio- Inspired strategies are found to be very useful in studying the PEG. While optimal response to the pursuit strategies are available using a geometric control theory, it is shown in this paper that application of linear feedback control laws can further improve the time and tracking response of these strategies in capturing the evader by the pursuer. Empirical results based on computer simulation are used to illustrate the findings. |
|
|
A new lane following approach based on deep learning with surround view images for an autonomous vehicle |
|
Minho Lee, Kyung Yeop Han, Jihun Yu, Seokhoon Ryu, Young-Sup Lee(Incheon National University, Korea) |
| Most lane following approaches use front camera which was attached to wind shield. In some cases, the lane lines may not be detected due to the limitation of front camera’s FOV (field of view). Some vehicles offer surround view images as one of the driver assistance systems. Therefore, in this paper, a lane following approach at high curvature road was implemented by using surround view images. Also, deep learning was used for robust lane line estimation for shadows and light reflections. |
|
|
An Iterative Learning Approach for Motion Control and Performance Enhancement of Quadcopter UAVs |
|
Mohammad Shaqura, Jeff S. Shamma(King Abdullah University of Science and Technology (KAUST), Saudi Arabia) |
| UAVs are agile systems that are often described with a nominal nonlinear model that neglects various complicated dynamic phenomena for the sake of easier analysis and control design. This simplification leads to limiting the vehicle performance. To overcome this issue, an iterative learning approach is presented where a nominal representation of the system dynamics is used in conjunction with flight trials to improve performance. The objective is to learn to aggressively navigate a quadcopter through a course while avoiding obstacles. |
|
|
Wall-contact sliding control strategy for a 2D caged quadrotor |
|
Pu Bai, Bruno J. N. Guerreiro, Rita Cunha(Instituto Superior Técnico, Universidade de Lisboa, Portugal), Przemyslaw Kornatowski, Dario Floreano(École Polytechnique Fédérale de Lausanne, Switzerland), Carlos Silvestre(University of Macau, Macao) |
| This paper addresses the trajectory tracking problem of a 2D caged flying robot in contact with a wall. The objective is to let the quadrotor hover or move along the wall with arbitrary attitude. The control law is derived using the Lyapunov stability theory, applying backstepping techniques to achieve exponential stability. To overcome unknown frictions between robot and wall, we design estimators for the friction coefficient, which include a projection operator that provides upper bounds for estimates. Realistic simulation results are provided to validate the proposed methodology. |
|