基于进化策略改进的D-S证据识别算法

Improved D-S Evidence Identification Algorithm Based on Evolutionar Strategies

提出了处理多传感器证据冲突的新方法,建立了获取传感器权重的优化模型,运用粒子群优化算法(PSO)求解传感器的权重分配,对Dempster-Shafter(D-S)证据理论进行改进。通过与其它改进算法的比较以及目标识别实例的仿真,验证了新方法的有效性和优越性。

The new method to deal with the evidence conflict of multi-sensor has been presented, and An optimize model of obtaining sensor weights has been set up. We use the particle swarm optimizaiton（PSO） to compute sensor weights to modify Dempster-Shafter（D- S） evidence theory. Through compared with other improved methods and a simulation example of target identification,the results prove the validity and superiority of new method.