Fig 1.
The modeled velocity to time curve of actual data, and the red line divides the v(t) curve into 3 parts.
Fig 2.
The sample of the traditional exponential speed model applied to our actual velocity data.
Fig 3.
Flowchart from data collection to the prediction performance comparison.
Fig 4.
Histogram of final time in our data.
It showed a right-skewed distribution.
Table 1.
Fig 5.
Description of the neural network model.
Only one hidden layer with 50 neurons using "rectified linear units" (ReLU).
Fig 6.
Description of one of the decision trees in the random forest model.
Fig 7.
Bar chart of the mean MSE for three methods.
Fig 8.
Two samples of NN model curve fitting.
Fig 8A is an example of better prediction with a lower MSE. Fig 8B shows an example of poor prediction performance. The red highlighted region represents the maximum phase of the NN modeled curve, including 0.98 * max speed to max speed.
Fig 9.
The predicted mean value of each 10m with 95% confidence interval.
Fig 10.
The scatter plot of each 10m with regression line.
The x-axis is the actual value and the y-axis is our predicted value.
Fig 11.
Correlation heatmap of maximum velocity, time of maximum velocity appearance, length of maximum velocity phase, and the final time.