基于改进复制动态演化博弈模型的行人与机动车冲突

针对现实社会中过街行人与机动车冲突的安全问题,本文根据人车双方冲突的实际背景，建立了人车冲突博弈矩阵，通过设置策略影响因子，给出策略之间的激励系数，对传统的演化复制动态方程进行改进，然后利用改进复制动态方程进行均衡求解。研究发现改进复制动态中激励系数的值不同，得到博弈演化的收敛速度不同。利用Matlab为实验平台，验证了策略之间既促进又抑制的关系对演化博弈过程的各种影响，突出了改进复制动态演化模型的合理性与优越性。为现实社会中过街行人与机动车冲突的安全问题提供了理论参考和依据。

Research on Network Attack and Defense Strategy Based on Evolutionary Game Model

Aiming at the safety problem of pedestrian vehicle conflict in the real society, this paper establishes the game matrix of pedestrian vehicle conflict according to the actual background of pedestrian vehicle conflict. By setting the influencing factors of the strategy, the incentive coefficient between the strategies is given, the traditional evolutionary replication dynamic process is improved, and then the improved replication dynamic equation is used for equilibrium solution. It is found that different values of the incentive coefficient in the improved replication dynamics give different convergence rates of game evolution. Using MATLAB as the experimental platform, the effects of promoting and restraining relations between strategies on the evolutionary game process are verified, and the rationality and superiority of improving the replication dynamic evolution model are highlighted. It provides a theoretical reference and basis for the safety problem of the conflict between pedestrians and motor vehicles in the real society.