In this paper, a hybrid approach organized in four phases is proposed to solve the multi-objective trajectory planning problem for industrial robots. In the first phase, a transcription of the original problem into a standard multi-objective parametric optimization problem is achieved by adopting an adequate parametrization scheme for the continuous robot configuration variables. Then, in the second phase, a global search is performed using a population-based search metaheuristic in order to build a first approximation of the Pareto front (PF). In the third phase, a local search is applied in the neighborhood of each solution of the PF approximation using a deterministic algorithm in order to generate new solutions. Finally, in the fourth phase, results of the global and local searches are gathered and postprocessed using a multi-objective direct search method to enhance the quality of compromise solutions and to converge toward the true optimal PF. By combining different optimization techniques, we intend not only to improve the overall search mechanism of the optimization strategy but also the resulting hybrid algorithm should keep the robustness of the population-based algorithm while enjoying the theoretical properties of convergence of the deterministic component. Also, the proposed approach is modular and flexible, and it can be implemented in different ways according to the applied techniques in the different phases. In this paper, we illustrate the efficiency of the hybrid framework by considering different techniques available in various numerical optimization libraries which are combined judiciously and tested on various case studies.

In the present work, a new hybrid approach combining particle swarm optimization (PSO) algorithm with recurrent dynamic neural network (RDNN), which is described as PSO-RDNN algorithm, is proposed for multi-performance optimization of machining parameters in finish turning of hardened AISI D2. The suggested optimization problem is solved using the weighted sum technique. Process parameters including cutting speed and feed rate are optimized for minimizing operation cost, maximizing tool life, and producing parts with acceptable surface roughness. Based on experimental results, two neural network models were developed for predicting tool flank wear and surface roughness during the machining process. Based on trained neural networks and structured hybrid algorithm, optimum cutting parameters were obtained. The coefficient of determination for trained neural networks was calculated as R2 = 0.9893 and R2 = 0.9879 for predicted flank wear and surface roughness, respectively, which proves the efficiency of trained neural models in real industrial applications. Furthermore, the offered methodology returns a Pareto optimality graph, which represents optimized cutting variables for several various cutting conditions.