Results 51 to 60 of about 16,456 (286)
In this study, a deep reinforcement learning framework with multi‐line‐of‐sight guidance is proposed. An augmented‐twin delayed deep deterministic policy gradient algorithm adapts to attitude changes efficiently, reducing computational costs. Simulations show 30.8% faster training and mitigates rudder/exergy issues, validated by an X‐rudder autonomous ...
Chengren Yuan+5 more
wiley +1 more source
Treecreeper Drone: Adaptive Mechanism for Passive Tree Trunk Perching
Taking inspiration from treecreepers, a passively triggered aerial robot that can reliably perch on vertical tree trunks is presented. The friction‐based approach combines a microspine array with a tail‐like support, and then validates via dynamic analyses and flight experiments, ensuring stable performance across trunk diameters and bark textures ...
Haichuan Li+2 more
wiley +1 more source
Dynamic multi‐objective optimisation of complex networks based on evolutionary computation
Abstract As the problems concerning the number of information to be optimised is increasing, the optimisation level is getting higher, the target information is more diversified, and the algorithms are becoming more complex; the traditional algorithms such as particle swarm and differential evolution are far from being able to deal with this situation ...
Linfeng Huang
wiley +1 more source
In this paper we investigate additional regularity properties for global and trajectory attractors of all globally defined weak solutions of semi-linear parabolic differential reaction-diffusion equations with discontinuous nonlinearities, when initial ...
Gluzman Mark O.+2 more
doaj +1 more source
Constrained Langevin approximation for the Togashi-Kaneko model of autocatalytic reactions
The Togashi Kaneko model (TK model) is a simple stochastic reaction network that displays discreteness-induced transitions between meta-stable patterns. Here we study a constrained Langevin approximation (CLA) of this model.
Wai-Tong (Louis) Fan+2 more
doaj +1 more source
Herein, a deep reinforcement learning‐based multi‐UAV formation control approach is proposed. By optimizing the utilization of historical data through correcting of offline samples, the past experience is better leveraged and learning performance is improved.
Zhongkai Chen+4 more
wiley +1 more source
Distributed wireless network resource optimisation method based on mobile edge computing
This paper mainly compares the network ranking leader, consumption amount and network signal reception of the three algorithms. The study found that in terms of network sort captain, there are significant differences between the CPLEX algorithm, the CCST algorithm, and edge computing methods. The CCST algorithm and edge computing have little difference
Jiongting Jiang+4 more
wiley +1 more source
Practical stability and Lyapunov functions [PDF]
The notion of "practical stability" was discussed in the monograph by LaSalle and Lefschetz [6] in which they point out that stability investigations may not assure "practical stability" and vice versa. For example an aircraft may oscillate around a mathematically unstable path, yet its performance may be acceptable.
Lakshmikantham, V., Bernfeld, Stephen R.
openaire +2 more sources
Abstract This article addresses the cooperative output consensus tracking problem for high‐order heterogeneous multi‐agent systems via a distributed proportional‐integral‐derivative (PID)‐like control strategy and proposes two novel control methodologies for the tuning of the control gains, which do not require any assumption and/or limitation on agent
Dario Giuseppe Lui+2 more
wiley +1 more source
This article describes an enhanced switched capacitor cross‐connected switched multilevel inverter (ESC3SMLI) with machine learning‐based model‐predictive control (ML‐MPCM). An improved switched capacitor cross connected switched multilevel inverter featuring minimal devices are presented. The proposed ESC3SMLI produces nine levels using eight switches,
Arun Vijayakumar+4 more
wiley +1 more source