Ontology-Driven Multi-Agent Engine for Real Time Adaptive Scheduling

In order to deal with the dynamic production environment with frequent fluctuation of processing time, robotic cell needs an efficient scheduling strategy which meets the real-time requirements. This paper proposes an adaptive scheduling method based on pattern classification algorithm to guide the online scheduling process. The method obtains the scheduling knowledge of manufacturing system from the production data and establishes an adaptive scheduler, which can adjust the scheduling rules according to the current production status. In the process of establishing scheduler, how to choose essential attributes is the main difficulty. In order to solve the low performance and low efficiency problem of embedded feature selection method, based on the application of Extreme Gradient Boosting model (XGBoost) to obtain the adaptive scheduler, an improved hybrid optimization algorithm which integrates Gini impurity of XGBoost model into Particle Swarm Optimization (PSO) is employed to acquire the optimal subset of features. The results based on simulated robotic cell system show that the proposed PSO-XGBoost algorithm outperforms existing pattern classification algorithms and the newly learned adaptive model can improve the basic dispatching rules. At the same time, it can meet the demand of real-time scheduling.

Download Full-text

A Geometric Approach to Dynamically Feasible, Real-Time Formation Control

Journal of Dynamic Systems Measurement and Control ◽

10.1115/1.4003209 ◽

2011 ◽

Vol 133 (2) ◽

Cited By ~ 7

Author(s):

D. H. A. Maithripala ◽

D. H. S. Maithripala ◽

S. Jayasuriya

Keyword(s):

Real Time ◽

Wide Class ◽

Formation Control ◽

Geometric Approach ◽

Control Problems ◽

Multi Agent Systems ◽

Individual Agent ◽

Planning Algorithm ◽

Explicit Consideration ◽

Multi Agent

We propose a framework for synthesizing real-time trajectories for a wide class of coordinating multi-agent systems. The class of problems considered is characterized by the ability to decompose a given formation objective into an equivalent set of lower dimensional problems. These include the so called radar deception problem and the formation control problems that fall under formation keeping and/or formation reconfiguration tasks. The decomposition makes the approach scalable, computationally economical, and decentralized. Most importantly, the designed trajectories are dynamically feasible, meaning that they maintain the formation while satisfying the nonholonomic and saturation type velocity and acceleration constraints of each individual agent. The main contributions of this paper are (i) explicit consideration of second order dynamics for agents, (ii) explicit consideration of nonholonomic and saturation type velocity and acceleration constraints, (iii) unification of a wide class of formation control problems, and (iv) development of a real-time, distributed, scalable, computationally economical motion planning algorithm.

Download Full-text