We present a genetic algorithm (GA) whose population possesses a spatial structure. The GA is formulated as a probabilistic cellular automaton: The individuals are distributed over a connected graph and the genetic op...
详细信息
We present a genetic algorithm (GA) whose population possesses a spatial structure. The GA is formulated as a probabilistic cellular automaton: The individuals are distributed over a connected graph and the genetic operators are applied locally in some neighborhood of each individual. By adding a self-organizing acceptance threshold schedule to the proportionate reproduction operator we can prove that the algorithm converges to the global optimum. First results for a multiple knapsack problem indicate a significant improvement in convergence behavior. The algorithm can be mapped easily onto parallel computers.
This paper presents a genetic-based approach to mobile robot motion planning with a distance-safety criterion. A wave front method is used to build the numerical potential fields for both the goal points and the obsta...
详细信息
This paper presents a genetic-based approach to mobile robot motion planning with a distance-safety criterion. A wave front method is used to build the numerical potential fields for both the goal points and the obstacles by representing the workspace as a grid. A computationally efficient genetic algorithm is proposed to search for near optimal paths, where a combined global and local optimization approach is employed to speed up the search process while considering the imposed requirements. Various simulation results show the effectiveness of the presented algorithm, including a comparison with the A* method.
This paper proposes a new method for speeding up geneticalgorithms (GAs) by temporarily protecting some components of chromosomes from ordinary genetic operators such as crossover and mutation. Protecting some compon...
详细信息
This paper proposes a new method for speeding up geneticalgorithms (GAs) by temporarily protecting some components of chromosomes from ordinary genetic operators such as crossover and mutation. Protecting some components of chromosomes can reduce the search space, speeding up GAs that, use this method. This new method is implemented on a subpopulation-type parallelized genetic algorithm. Each subpopulation protects some promising components of its chromosomes, and improves those components in cooperation with or competition against other subpopulations. Common components of chromosomes in a subpopulation are chosen as a protected part in that subpopulation, for convenience. Several travelling salesman problems are used to illustrate the accelerating effect of this method. The simulation results show that a GA using this method is able to search about twice as fast as a non-protecting parallelized GA.
This paper presents an application of geneticalgorithms for dynamic lotsizing problems, including the implementation methodology and the testing results of the algorithms. Currently most of the existing studies for d...
详细信息
This paper presents an application of geneticalgorithms for dynamic lotsizing problems, including the implementation methodology and the testing results of the algorithms. Currently most of the existing studies for dynamic lotsizing problems concentrate on heuristic lot-sizing techniques which only consider some simple production structures acid or simple external demands structures. In this paper, the general dynamic lot-sizing problems are considered, which are characterized by the fact that each stage may have several predecessor and/or successor stages, all the items can have independent requirements, and/or all the cost parameters can be time-varying. A genetic algorithm for the problems is introduced, which attempt to heuristically optimize under all the conditions simultaneously. As to my knowledge, this genetic algorithm is the first one capable of solving such general dynamic lotsizing problems. In order to apply genetic algorithm, a coding scheme for lotsize plan/schedule is given and a feasibility routine is presented. In computational experiments, this genetic algorithm performed extremely well. It is concluded that the genetic algorithm is efficient and effective for dynamic lotsizing problems.
This paper presents the use of geneticalgorithms as a search method for Inverse Synthetic Aperture Radar (ISAR) image focusing. Although motion estimation using joint time-frequency representations allows much of the...
详细信息
This paper presents the use of geneticalgorithms as a search method for Inverse Synthetic Aperture Radar (ISAR) image focusing. Although motion estimation using joint time-frequency representations allows much of the blurring to be removed, there remain noticeable effects due to imperfect compensation. The genetic Algorithm is used as a fine tuning process in which a measure of focus is used as the fitness function corresponding to motion parameter values which make up each gene. ISAR image examples from simulated and real data are shown.
The conventional mechanism used to gain fault tolerance is redundancy. In contrast, this paper suggests that artificial evolution can be used to produce systems that are inherently insensitive to faults, with fault to...
详细信息
The conventional mechanism used to gain fault tolerance is redundancy. In contrast, this paper suggests that artificial evolution can be used to produce systems that are inherently insensitive to faults, with fault tolerance becoming part of the task specification. The possible techniques are investigated, and the study is grounded in a real-world evolved electronic control system for a robot.
This paper discusses `the shape of space', in terms of search algorithms and traversal operators. We point out that it is the combination of representation and traversal operators that defines an algorithm's v...
详细信息
This paper discusses `the shape of space', in terms of search algorithms and traversal operators. We point out that it is the combination of representation and traversal operators that defines an algorithm's view of a given search problem, and hence gives rise to a fitness landscape. We provide an intuitive background to some recent formal discussions on the limitations of search algorithms, and demonstrate how these issues arise in geneticalgorithms (GAs) and encoded stochastic hill-climbers. We suggest that randomly re-mapping space via base changes provides a simple means of applying multiple search strategics to a given search problem, and that this offers a pragmatic means for probing a cost function from many views. We introduce a number of new algorithms based on this technique and demonstrate their application on a range of standard cost functions.
The paper presents a novel two-layer learning method for radial basis function (RBF) networks. At the lower layer, a regularized orthogonal least squares (ROLS) algorithm is employed to construct RBF networks while th...
详细信息
The paper presents a novel two-layer learning method for radial basis function (RBF) networks. At the lower layer, a regularized orthogonal least squares (ROLS) algorithm is employed to construct RBF networks while the two key learning parameters, the regularization parameter and hidden node's width, needed by the ROLS algorithm are optimized using the genetic algorithm at the higher layer. Networks constructed by this learning method have superior generalization properties, and the computational complexity of the method is reasonable. Nonlinear time series modelling and prediction is used as an example to demonstrate the effectiveness of this hierarchical learning approach.
This paper develops high performance system identification and linearization techniques, using a genetic algorithm. The algorithm is fine tuned by simulated annealing, which yields a faster convergence and a more accu...
详细信息
This paper develops high performance system identification and linearization techniques, using a genetic algorithm. The algorithm is fine tuned by simulated annealing, which yields a faster convergence and a more accurate search. This global search technique is used to identify the parameters of a system described by an ARMAX model in the presence of white noise and to approximate a nonlinear multivariable system by a linear time-invariant state space model. Results obtained show that simple step input can be used for effective system identification and linearization with much higher performance than conventional means.
geneticalgorithms (GAs) have been designed as general purpose optimization methods. GAs can be uniquely characterized by their population-based search strategies and their operators: mutation, selection and crossover...
详细信息
geneticalgorithms (GAs) have been designed as general purpose optimization methods. GAs can be uniquely characterized by their population-based search strategies and their operators: mutation, selection and crossover. In this paper, we propose a new crossover called multi-step crossover (MSX) which utilizes a neighborhood structure and a distance in the problem space. Given parents, MSX successively generates their descendents along the path connecting the both of them. MSX was applied to the job-shop scheduling problem (JSSP) as a high-level crossover to work on the critical path. Preliminary experiments using JSSP benchmarks showed the promising performance of a GA with the proposed MSX.
暂无评论