fig6

Optimal consensus control of Markov jump multi-agent systems under switching topology

Figure 6. Iterative error in GPI and GVI algorithms. GPI: Global policy iteration; GVI: global value iteration.