Xiaohui Hu

Multiobjective optimization using dynamic neighborhood particle swarm optimization

This paper presents a particle swarm optimization (PSO) algorithm for multiobjective optimization problems. PSO is modified by using a dynamic neighborhood strategy, new particle memory updating, and one-dimension optimization to deal with multiple objectives. Several benchmark cases were tested and showed that PSO could efficiently find multiple Pareto optimal solutions.

Recent advances in particle swarm

This paper reviews the development of the particle swarm optimization method in recent years. Included are brief discussions of various parameters. Modifications to adapt to different and complex environments are reviewed, and real world applications are listed.

Adaptive particle swarm optimization: detection and response to dynamic systems

This paper introduces an adaptive PSO, which automatically tracks various changes in a dynamic system. Different environment detection and response techniques are tested on the parabolic and Rosenbrock benchmark functions, and re-randomization is introduced to respond to the dynamic changes. Performance on the benchmark functions with various severities is analyzed.

Swarm intelligence for permutation optimization: a case study of n-queens problem

This paper introduces a modified particle swarm optimizer which deals with permutation problems. Particles are defined as permutations of a group of unique values. Velocity updates are redefined based on the similarity of two particles. Particles change their permutations with a random rate defined by their velocities. A mutation factor is introduced to prevent the current pBest from becoming stuck at local minima. Preliminary study on the n-queens problem shows that the modified PSO is promising in solving constraint satisfaction problems.

An analysis of Bare Bones Particle Swarm

The bare bones particle swarm (BBPS) is evolved from the canonical particle swarm optimizer (PSO). The velocity term of the canonical PSO is removed in BBPS and replaced by Gaussian sampling strategy. There is no parameter tuning and it is much easier to implement. In the paper, it is proven that the BBPS can be mathematically deduced from the canonical PSO and a more general formula of BBPS is also presented. The results presented in the paper represent initial results of an ongoing research project effort.

A new UAV assignment model based on PSO

An unmanned aerial vehicle (UAV) assignment model requires allocating vehicles to targets to perform various tasks. It is a complex assignment problem with hard constraints, and potential dimensional explosion when the scenarios become more complicated and the size of problems increases. In this paper, a new UAV assignment model is proposed which reduces the dimension of the solution space and can be easily adapted by computational intelligence algorithms. In the proposed model a local version of particle swarm optimization (PSO) is applied to accomplish the optimization work. Numerical experimental results illustrate that it can efficiently achieve the optima and demonstrate the effectiveness of combining the model and a local version of PSO to solve complex UAV assignment problems.

Human vs. swarm: An NK landscape game

This paper describes a computer game the purpose of which is to investigate how humans interact with swarm intelligence. The game is based on an NK landscape as described by Stuart Kaufmann. It is concluded that the combination of a human-swarm team may have advantages in certain environments, such as dynamic decision making tasks. The team approach can combine computer computational power with human intuitive knowledge to provide improved performance for dynamic and complex tasks.

Modeling drowsy driving behaviors

Excessive sleepiness may result in an increased risk of a motor vehicle crash either because the motorist falls asleep while driving or because he/she experiences reduced attention to road events and driving tasks due to sleepiness/inattention. This study was designed to investigate noninvasive measurable patterns that predict driving-related sleepiness and inattention. Seventeen residents-in-training (residents) recruited from the Indiana University Hospital took five-session driving tests on a driving simulator. Driving, sleep diary, questionnaire, and electroencephalogram (EEG) information were recorded for subsequent data analysis. With statistical and computational intelligence tools, some basic driving behaviors associated with inattention and sleepiness were identified. Results suggest that a combination of standard deviation of lateral lane position and standard deviation of steering wheel angle is a possible measure of the relationships between driving behaviors and road risks. The derived patterns are also consistent with other non-invasive measurements of sleepiness such as Epworth Sleepiness Scale and Stanford Sleepiness Scale.