An Intelligent Variable Spraying Decision-Making System Based on Fuzzy Neural Network for Greenhouse Mobile Robot - Intelligent Computing for Sustainable Energy and Environment - page 263

Tabl e 1. Part of Sample Data

Sample
Number

Area
( m 2 )

Distance
( m )

Damage
level

Pesticide
Quantity
( L · s − 1 )

1 0.00656 0.2394 0 0.02125

2 0.00689 0.2315 0 0.02131

3 0.00736 0.2382 0 0.02370

4 0.02074 0.2458 0 0.02573

5 0.02579 0.3367 0 0.02948

6 0.02584 0.4285 0 0.0542

7 0.05891 0.2759 0 0.02917

8 0.08652 0.3756 0 0.05341

9 0.09010 0.4152 0 0.06324

10 0.01949 0.2521 1 0.02932

11 0.02198 0.3516 1 0.07301

12 0.01736 0.4156 1 0.05629

13 0.03403 0.2657 1 0.05438

14 0.01863 0.2981 1 0.04730

15 0.04517 0.4162 1 0.07307

16 0.06086 0.2429 1 0.04833

17 0.06103 0.3507 1 0.05924

18 0.09211 0.4128 1 0.06269

Validate the decision-making precision of the trained fuzzy neural network
with the reserved test samples in the spraying mobile robot system. The error
between the actual output and the sample output is shown in Fig. 6. In the
same condition, the fuzzy decision system is simulated by using the same fuzzy
division method and reserved test sample. The error is shown in Fig. 6.

Performance is 0.000996019, Goal is 0.001

10 0

10 −1

10 −2

10 −3

10 −4

0 50 100 150 200 250

191 Epochs

Fig. 4. Results of the network learning