| 1. |
- Jang, Dohyun, et al.
(author)
-
Efficient networked UAV control using event-triggered predictive control
- 2019
-
In: IFAC PAPERSONLINE. - : ELSEVIER. - 2405-8963. ; , s. 412-417
-
Conference paper (peer-reviewed)abstract
- In this paper, we propose a method to improve the networked UAV control system using event-triggered control and model predictive control (MPC). Although the UAV control over the network has many advantages, it involves a long-time delay and packet loss, which adversely affect real-time control performance. Delay compensation algorithms in the networked control system (NCS) have been proposed to address such issues, however, they do not consider the resource limit of the network so that the network congestion may occur. In that case, the packet loss and network delay issues can even be worsened. In this study, we propose a method to reduce the generation of less important control signals and to use the network more efficiently by using event-triggered control. Since the event-triggered control method is also influenced by the network delay, an event trigger function suitable for NCS is designed. We validated the effectiveness of networked UAV control system and event-triggered control by simulation.
|
|
| 2. |
- Lee, Seung Jae, et al.
(author)
-
Design, Modeling and Control of T-3-Multirotor : a Tilting Thruster Type Multirotor
- 2018
-
In: 2018 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA). - : IEEE Computer Society. - 9781538630815 ; , s. 1214-1219
-
Conference paper (peer-reviewed)abstract
- This paper presents a new design of multirotor, named as 'Tilting Thruster Type' (T-3)-multirotor. The new platform is equipped with mechanically separated thrusters, which can take any fuselage posture within a specified range regardless of any direction of translational acceleration. A specially designed servo-linkage mechanism is employed for relative attitude control between the thruster and the fuselage. Mathematical modeling and analysis of the new platform are conducted to explore the control method of the dynamically complex system. For demonstrating the potential of the new T(3-)multirotor, an autonomous level flight is performed where the fuselage maintains zero roll and pitch angle during the entire flight. Both simulation and experimental results are provided with detailed analysis.
|
|
| 3. |
- Yoo, Jaehyun, et al.
(author)
-
Event-based Observer and MPC with Disturbance Attenuation using ERM Learning
- 2018
-
In: 2018 European Control Conference, ECC 2018. - : Institute of Electrical and Electronics Engineers (IEEE). - 9783952426982 ; , s. 1894-1899
-
Conference paper (peer-reviewed)abstract
- This paper presents a learning-based approach for disturbance attenuation for a non-linear dynamical system with event-based observer and model predictive control (MPC). Using the empirical risk minimization (ERM) method, we can obtain a learning error bound which is function of the number of samples, learning parameters, and model complexity. It enables us to analyze the closed-loop stability in terms of the learning property, where the state estimation error by the ERM learning is guaranteed to be bounded. Simulation results underline the learning's capability, the control performance and the event-triggering efficiency in comparison to the conventional event-triggered control scheme.
|
|
| 4. |
- Yoo, Jaehyun, et al.
(author)
-
Event-Triggered Model Predictive Control With a Statistical Learning
- 2021
-
In: IEEE Transactions on Systems, Man & Cybernetics. Systems. - : Institute of Electrical and Electronics Engineers (IEEE). - 2168-2216 .- 2168-2232. ; 51:4, s. 2571-2581
-
Journal article (peer-reviewed)abstract
- The event-triggered model predictive control (MPC) reduces energy consumption for updating control sequences while maintaining the originality of the MPC, which copes with hard constraints on dynamical systems. In the presence of large uncertainties, however, the standard event-triggered MPC may generate too frequent event occurrences. To compensate for unknown uncertainties, this paper applies a statistical learning to event-triggered MPC. The stability and the feasibility of the proposed control system are analyzed in regard to the statistical learning influences, such as the number of training samples, model complexity, and learning parameters. Accordingly, the event-triggering policy is established to guarantee the stability. We evaluate the proposed algorithm for the tracking problem of a nonholonomic model perturbed by uncertainties. In comparison with the standard event-triggered control scheme, the simulation results of the proposed method show better tracking performance with less frequent event triggers.
|
|
| 5. |
- Yoo, Jaehyun, et al.
(author)
-
Event-triggered Model Predictive Control with Machine Learning for Compensation of Model Uncertainties
- 2017
-
In: 2017 IEEE 56th Annual Conference on Decision and Control, CDC 2017. - : Institute of Electrical and Electronics Engineers (IEEE). - 9781509028733 ; , s. 5463-5468
-
Conference paper (peer-reviewed)abstract
- As one of the extensions of model predictive control (MPC), event-triggered MPC takes advantage of the reduction of control updates. However, approaches to event-triggered MPCs may be subject to frequent event-triggering instants in the presence of large disturbances. Motivated by this, this paper suggests an application of machine learning to this control method in order to learn a compensation model for disturbance attenuation. The suggested method improves both event-triggering policy efficiency and control accuracy compared to previous approaches to event-triggered MPCs. We employ the radial basis function (RBF) kernel based machine learning technique. By the universial approximation property of the RBF, which imposes an upper bound on the training error, we can present the stability analysis of the learningaided control system. The proposed algorithm is evaluated by means of position control of a nonholonomic robot subject to state-dependent disturbances. Simulation results show that the developed method yields not only two times less event triggering instants, but also improved tracking performance.
|
|
| 6. |
|
|
| 7. |
- Yoo, Jaehyun, et al.
(author)
-
Learning communication delay patterns for remotely controlled UAV networks
- 2017
-
In: IFAC PAPERSONLINE. - : Elsevier. ; , s. 13216-13221
-
Conference paper (peer-reviewed)abstract
- This paper deals with collaborative unmanned aerial vehicles (UAVs) that are remotely controlled from a cloud server. The main contribution is to apply machine learning technique to find a pattern of network-induced effects on maneuvers of UAVs, in order to compensate for time delays and packet losses in remote communication. As machine learning technique, a Gaussian process (GP) based approach is employed due to its computational simplicity and flexibility in modelling complex expressions using a small number of parameters. We combine a deterministic compensation for an enhanced GP model to overcome a problem of the lack of training data at the beginning of training phase. This is done by defining training data input as a set of delayed observation and the deterministic compensation term, and by training the GP on residual between the true state and the input set. The proposed algorithm is evaluated to collaborative trajectory tracking of two UAVs. Simulations are performed for various delays and tracking scenarios. It is shown that the better tracking results are achieved compared to a conventional linear compensation algorithm.
|
|
| 8. |
- Yoo, Jaehyun, et al.
(author)
-
Mapless Indoor Localization by Trajectory Learning from a Crowd
- 2016
-
In: 2016 INTERNATIONAL CONFERENCE ON INDOOR POSITIONING AND INDOOR NAVIGATION (IPIN). - : Institute of Electrical and Electronics Engineers (IEEE). - 9781509024254
-
Conference paper (peer-reviewed)abstract
- This paper suggests a mapless indoor localization using wifi received signal strength (RSS) of a smartphone, collected by multiple people. A new trajectory learning algorithm by combining a dynamic time warping and a machine learning technique is proposed in order to generate an alternative map. Moreover, we combine particle filter and Gaussian process (GP) for the position estimation, because it can use the alternative map as the probabilistic function (the prior), and can use probabilistic relationship (the likelihood) between wifi RSSs and location. Field experimental results confirm the usefulness of our algorithm when the map is not available and robustness against outliers, in that the accuracy of the proposed localization is similar to that using the true map information.
|
|
| 9. |
- Yoo, Jaehyun, et al.
(author)
-
Path planning for remotely controlled UAVs using Gaussian process filter
- 2017
-
In: 17th International Conference on Control, Automation and Systems, ICCAS 2017. - : IEEE Computer Society. - 9788993215137 ; , s. 477-482
-
Conference paper (peer-reviewed)abstract
- Most of the recent results in control of unmanned aerial vehicles (UAVs) have focused on motion stability and navigation in well-structured environments, without considering communication delay influences. In order to deal with time delays and packet losses in networked UAVs, this paper suggests a machine learning based Gaussian process (GP) filter for a path planning problem. The developed GP filter estimates the UAV states accurately given delayed observation by learning the pattern of network-induced effects on UAV maneuvers. We validate that the GP filter produces the lower error rate than Kalman filter by analyzing error covariances. The proposed algorithm is evaluated on a collaborative trajectory tracking task for two networked-UAVs and the better control performance is achieved.
|
|
| 10. |
- Yoo, Jaehyun, et al.
(author)
-
Semi-Supervised Learning for Mobile Robot Localization using Wireless Signal Strengths
- 2017
-
In: 2017 international conference on indoor positioning and indoor navigation (ipin). - 9781509062997
-
Conference paper (peer-reviewed)abstract
- This paper proposes a new semi-supervised machine learning for localization. It improves localization efficiency by reducing efforts needed to calibrate labeled training data by using unlabeled data, where training data come from received signal strengths of a wireless communication link. The main idea is to treat training data as spatio-temporal data. We compare the proposed algorithm with the state-of-art semi-supervised learning methods. The algorithms are evaluated for estimating the unknown location of a smartphone mobile robot. The experimental results show that the developed learning algorithm is the most accurate and robust to the varying amount of training data, without sacrificing the computation speed.
|
|
