提出一种空间联合概率数据关联的多目标粒子群优化(DS-MOPSO)算法.采用正态分布确保初始样本均匀分布,通过采用拥挤距离和先验概率采样确立外部归档中非支配解的拥挤度来保持解的多样性;采用Sigma方法作为选择精英粒子策略寻找全局最优解;利用空间联合概率数据关联动态生成每个粒子的惯性权值,增强粒子的搜索区域,防止算法陷入局部最优.仿真实验结果表明,采用所提出的算法所得到的Pareto解集具有很好的收敛性和多样性. This paper proposes a multi-objective Particle Swarm Optimization (DS-MOPSO) algorithm for spatial joint probability data association, which ensures the uniform distribution of initial samples by using normal distribution and establishes the crowding degree of non-dominated solutions in external archive by using crowding distance and prior probability sampling To maintain the diversity of solutions; Sigma method is used as the elitist particle selection strategy to find the global optimal solution; the spatial joint probability data is used to dynamically generate the inertia weight of each particle to enhance the particle search area and prevent the algorithm from falling into the local optimum. Simulation results show that the Pareto solution set obtained by the proposed algorithm has good convergence and diversity.