Advances, Systems and Applications
From: Reinforcement learning empowered multi-AGV offloading scheduling in edge-cloud IIoT
Notation | Description |
---|---|
M, m | Total number of AGVs, m \(\in \{1,2,3,...,M\}\) |
N, n | Total number of ESes, n \(\in \{1,2,3,...,N\}\) |
v | Driving speed of AGVs |
\(L_{total}^m,L_{i,j}^m\) | The distance traveled by the m-th AGV in one cycle and the distance from point i to point j |
\({Task}^m\) | Total number of tasks carried by the m-th AGV in one cycle |
\(T_m\) | Total time spent by the m-th AGV in one cycle including loading/offloading tasks and traveling |
\(t_m^{on}\) | The time spent by the m-th AGV to load tasks at the starting point |
\(t_m^{mov}\) | Travel time of the m-th AGV |
\(t_m^{off}\) | Task offloading time of the m-th AGV |
\(I_m\) | Starting point of the m-th AGV |
\(D_m\) | End point of the m-th AGV |
\(|D_m-I_m|\) | Number of ESes that the m-th AGV passes from the beginning to the end |
\(C_n\) | The capacity of n-th ES |
\(E_n\) | Processing efficiency of n-th ES |
\({Task}^m_i\) | Number of tasks offloaded by the m-th AGV at the i-th ES |
\(C_i^m\) | Available capacity of the i-th ES when the m-th AGV passes through the i-th ES |
\({RT}_i^m\) | Number of tasks allocated on the i-th ES for the remaining tasks of the m-th AGV |
\({RC}_i^m\) | Remaining available resources of the i-th ES in the m-th AGV path |