tailieunhanh - Optimal robotic assembly sequence generation using particle swarm optimization

In this methodology, each component of the assembled product is considered as the particle (bird) and mutation operation is performed to generate a new assembly sequence for each iteration. To generate optimal assembly sequence, a fitness function is generated, which is based on the energy function and robot directional changes associated with assembly sequence. | Journal of Automation and Control Engineering Vol. 4, No. 2, April 2016 Optimal Robotic Assembly Sequence Generation Using Particle Swarm Optimization M. V. A. Raju. Bahubalendruni, B. B. Biswal, and B. B. V. L Deepak Dept. of Industrial Design, National Institute of Technology- Rourkela, Odisha-769008, India Email: first. bahubalindruni@, {bbbiswal, bbv}@ this fact researcher’s interest is growing in this field. An important aspect of this developing process is represented by the need to automatically generate the assembly plan by identifying the optimum sequence of operations with respect to its cost and correctness. Products with large number of parts have several alternative feasible sequences among which optimal assembly sequence is generated. Baldwin et al. [1] developed simplified method which can find the optimal solutions, but have a problem of the search space explosion for an increased number of parts. Hong and Cho [2]-[4] proposed neural-network based computational approaches, which have been reported to overcome the problem of the search space explosion. However, the methods have a problem of frequent generation of no optimal sequences, since the network energy often reaches to a local minimum. Cho and Cho [5] developed a method using directional part contact level graphs which contains the information on directional connections for each pair of mating parts. Lee [6] proposed disassembly method. In this method, an assembly sequence was determined by the reverse order of disassembly sequence expressed in a list of parts each of which is sequentially chosen to have minimum cost of disassembly. These are some of the classical approaches for solving assembly sequencing plan. Besides the above mentioned techniques, researchers have also concentrated on artificial intelligence techniques for solving the same problem but with less mathematical complexity. Wang et al. [7] proposed ant algorithm by using the disassembly operations of .