ST-AGRNN: A Spatio-Temporal Attention-Gated Recurrent Neural Network for Traffic State Forecasting

<table class="figure-group"><tr class="fig-image" id="a"><td><object data="https://static-preview.hindawi.com/articles/jat/volume-2022/2806183/figures/2806183.fig.004a.svgz" name="2806183.fig.004a" type="image/svg+xml"></object></td></tr><tr class="fig-caption"><td><b>(a) </b>Node 111 in PeMSD4 for 15 min</td></tr><tr class="fig-image" id="b"><td><object data="https://static-preview.hindawi.com/articles/jat/volume-2022/2806183/figures/2806183.fig.004b.svgz" name="2806183.fig.004b" type="image/svg+xml"></object></td></tr><tr class="fig-caption"><td><b>(b) </b>Node 261 in PeMSD4 for 15 min</td></tr><tr class="fig-image" id="c"><td><object data="https://static-preview.hindawi.com/articles/jat/volume-2022/2806183/figures/2806183.fig.004c.svgz" name="2806183.fig.004c" type="image/svg+xml"></object></td></tr><tr class="fig-caption"><td><b>(c) </b>Node 111 in PeMSD4 for 60 min</td></tr><tr class="fig-image" id="d"><td><object data="https://static-preview.hindawi.com/articles/jat/volume-2022/2806183/figures/2806183.fig.004d.svgz" name="2806183.fig.004d" type="image/svg+xml"></object></td></tr><tr class="fig-caption"><td><b>(d) </b>Node 261 in PeMSD4 for 60 min</td></tr></table>

<div>Traffic flow forecast visualization in PeMSD4.</div>

Journal of Advanced Transportation

fig4

Figure 4

Figure 4: ST-AGRNN: A Spatio-Temporal Attention-Gated Recurrent Neural Network for Traffic State Forecasting 