Using a mathematical model of a quadrotor, the Euler angle dynamics equations are analyzed to obtain a linear regression that describes the plant parametric uncertainty. Based on this representation, model reference a...
详细信息
Deep learning methods have been widely used in robotic applications, making learning-enabled control design for complex nonlinear systems a promising direction. Although deep reinforcement learning methods have demons...
详细信息
ISBN:
(纸本)9798350382662;9798350382655
Deep learning methods have been widely used in robotic applications, making learning-enabled control design for complex nonlinear systems a promising direction. Although deep reinforcement learning methods have demonstrated impressive empirical performance, they lack the stability guarantees that are important in safety-critical situations. One way to provide these guarantees is to learn Lyapunov certificates alongside control policies. There are three related problems: 1) verify that a given Lyapunov function candidate satisfies the conditions for a given controller on a region, 2) find a valid Lyapunov function and controller on a given region, and 3) find a valid Lyapunov function and a controller such that the region of attraction is as large as possible. Previous work has shown that if the dynamics are piecewise linear, it is possible to solve problems 1) and 2) by solving a Mixed-Integer Linear Program (MILP). In this work, we build upon this method by proposing a Lyapunov neural network that considers monotonicity over half spaces in different directions. We 1) propose a specific choice of Lyapunov function architecture that ensures non-negativity and a unique global minimum by construction, and 2) show that this can be leveraged to find the controller and Lyapunov certificates faster and with a larger valid region by maximizing the size of a square inscribed in a given level set. We apply our method to a 2D inverted pendulum, unicycle path following, a 3-D feedback system, and a 4-D cart pole system, and demonstrate it can shorten the training time by half compared to the baseline, as well as find a larger ROA.
The road gradient presents a significant challenge and offers crucial insights for vehicle control, profoundly influencing drivability. Numerous studies conventionally derive road gradient data in advance from high-pr...
详细信息
This study investigates the impact of utilizing a network RTK and our own RTK base station on the lateral tracking performance of a GNSS-based navigation system. The experiment employs an adaptive PID controller and c...
详细信息
As modern large cities heavily rely on urban rail transit (URT) systems as their most crucial transportation infrastructure to meet daily commuting demands, the irrational design of the first train timetable and the l...
详细信息
This paper presents a holistic approach for transmitting arbitrary data to local Internet-of-Things (IoT) devices using only public services in a zero-trust environment. The approach features a variety of benefits, na...
详细信息
ISBN:
(纸本)9798350371000;9798350370997
This paper presents a holistic approach for transmitting arbitrary data to local Internet-of-Things (IoT) devices using only public services in a zero-trust environment. The approach features a variety of benefits, namely (i) eliminates the need for service providers to deploy their own costly infrastructure, (ii) enables the receiving device to operate without an active network connection, (iii) ensures that the data can be proven untampered on the IoT device itself, (iv) verifies both the validity and recency of the information, (v) requires low administrative and technical efforts, (vi) allows for wireless data updates, including the timeliness of current information, by any participant due to the zero-trust assumption. The described methodology utilizes distinct features of Proof-of-Work (PoW)-based programmable blockchain systems. This work will focus on the usage of Ethereum Classic (ETC) as base layer for trust, the validity and recency of the information. The proposed approach has wide-ranging practical applications, including programmable, rule-based locking systems such as smart locks used in corporate and institutional buildings, cars, and hotels. It also facilitates other forms of access control, such as ticket systems, health certificates, and proof-of-possession for resources or achievements. Due to the zero-trust assumption, data updates as well their recency can be updated by any participant of the system.
With wireless technology is widely used in outdoor lighting controlsystems, the wireless networkperformance determines the system operation quality. The lighting pole distance and antenna height influence the wirele...
详细信息
This study proposes a data-driven controller design for Adaptive Cruise control (ACC) systems. The design employs Long Short-Term Memory (LSTM) networks to enable dynamic customisation and superior performance over cl...
详细信息
A multi-hop wireless network forms a formidable solution to overcome spectrum deficit and coverage issues in wireless conditions. As wireless internet remains crucial in connecting people from remote locations, a vast...
详细信息
We proposed the Sparse Bayesian network-based Disturbance Observer (SBN-DOB) to enhance the robustness of policy-based reinforcement learning. SBN-DOB utilizes sparse Bayesian learning to estimate the nominal inverse ...
详细信息
暂无评论