Dynamic Task Planning for Multi-Arm Harvesting Robots Under Multiple Constraints Using Deep Reinforcement Learning

Horticulturae, Journal Year: 2025, Volume and Issue: 11(1), P. 88 - 88

Published: Jan. 14, 2025

Global fruit production costs are increasing amid intensified labor shortages, driving heightened interest in robotic harvesting technologies. Although multi-arm coordination robots is considered a highly promising solution to this issue, it introduces technical challenges achieving effective coordination. These include mutual interference among mechanical structures, task allocation across multiple arms, and dynamic operating conditions. This imposes higher demands on for robots, requiring collision-free collaboration, optimization of sequences, re-planning. In work, we propose framework that models the planning problem operation as Markov game. First, considering cooperative movement picking sequence optimization, employ two-agent game model robot problem. Second, introduce self-attention mechanism centralized training execution strategy design our deep reinforcement learning (DRL) model, thereby enhancing model’s adaptability uncertain environments improving decision accuracy. Finally, conduct extensive numerical simulations static environments; when targets set 25 50, time reduced by 10.7% 3.1%, respectively, compared traditional methods. Additionally, environments, both operational efficiency robustness superior approaches. The results underscore potential approach revolutionize robotics providing more adaptive efficient solution. We will research positioning accuracy fruits future, which make possible apply real robots.

Language: Английский

---

An advantage duPLEX dueling multi-agent Q-learning algorithm for multi-UAV cooperative target search in unknown environments

Simulation Modelling Practice and Theory, Journal Year: 2025, Volume and Issue: unknown, P. 103118 - 103118

Published: April 1, 2025

Language: Английский

---

Simulation Model as an Element of Sustainable Autonomous Mobile Robot Fleet Management

Energies, Journal Year: 2025, Volume and Issue: 18(8), P. 1894 - 1894

Published: April 8, 2025

Computer simulations of processes are increasingly used in business practice to improve the results an enterprise and maximise its value. Designing process models simulating their behaviour provide opportunity analyse economic operational before appropriate organisational, location, investment decisions made. This article presents possibilities using simulation modelling intralogistics systems. In presented article, a decision-making support tool based on DES simulator developed by authors was proposed. supports analysis parameters that affect energy efficiency analysed sustainability. The proposed were giving example implementation automation processes. As part implementation, use Autonomous Mobile Robot (AMR) vehicles By conducting experiments system model analysing obtained also terms consumption AMR vehicles, project can be verified improvements this research confirmed possibility for supporting assessing designed system. method is cost-free element helps management staff given make decisions.

Language: Английский

---

Intelligent Scheduling Methods for Optimisation of Job Shop Scheduling Problems in the Manufacturing Sector: A Systematic Review

Electronics, Journal Year: 2025, Volume and Issue: 14(8), P. 1663 - 1663

Published: April 19, 2025

This article aims to review the industrial applications of AI-based intelligent system algorithms in manufacturing sector find latest methods used for sustainability and optimisation. In contrast previous articles that broadly summarised existing methods, this paper specifically emphasises most recent techniques, providing a systematic structured evaluation their practical within sector. The primary objective study is algorithms, including metaheuristics, evolutionary learning-based sector, particularly through lens optimisation workflow production lines, Job Shop Scheduling Problems (JSSPs). It critically evaluates various solving JSSPs, with particular focus on Flexible (FJSPs), more advanced form JSSPs. process consists several intricate operations must be meticulously planned scheduled executed effectively. regard, Production scheduling best possible schedule maximise one or performance parameters. An integral part JSSP both traditional smart manufacturing; however, research focuses concept general, which pertains concerns aim maximising operational efficiency by reducing time costs. A common feature among studies lack consistent effective solution minimise energy consumption, thus accelerating minimal resources.

Language: Английский

---

A framework proposal for scheduling environmental impact evaluation in manufacturing systems

International Journal of Production Research, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 19

Published: May 14, 2025

Language: Английский

---

An Optimization Method for Green Permutation Flow Shop Scheduling Based on Deep Reinforcement Learning and MOEA/D

Machines, Journal Year: 2024, Volume and Issue: 12(10), P. 721 - 721

Published: Oct. 11, 2024

This paper addresses the green permutation flow shop scheduling problem (GPFSP) with energy consumption consideration, aiming to minimize maximum completion time and total as optimization objectives, proposes a new method that integrates end-to-end deep reinforcement learning (DRL) multi-objective evolutionary algorithm based on decomposition (MOEA/D), termed GDRL-MOEA/D. To improve quality of solutions, study first employs DRL model PFSP sequence-to-sequence (DRL-PFSP) obtain relatively better solutions. Subsequently, solutions generated by DRL-PFSP are used initial population for MOEA/D, proposed job postponement energy-saving strategy is incorporated enhance solution effectiveness MOEA/D. Finally, comparing GDRL-MOEA/D NSGA-II, marine predators (MPA), sparrow search (SSA), artificial hummingbird (AHA), seagull (SOA) through experimental tests, results demonstrate has significant advantage in terms quality.

Language: Английский

---

Genetic and Deep Reinforcement Learning-Based Intelligent Course Scheduling for Smart Education

Published: Oct. 12, 2024

Language: Английский

---