LoRa Communications as an Enabler for Internet of Drones towards Large-Scale Livestock Monitoring in Rural Farms

a

) (

b

) 

Figure 19. 

(

a

) The optimal flight path calculated by the PSO algorithm and (

b

) the performance of the PSO algorithm under 

different sizes of the swarm. 

To solve the challenge, the idea of mutation from the GA was used and applied to 

the TSP. Therefore, in each iteration, the EPSO algorithm generates some random new 

solutions for both personal and global cases, resulting in a higher exploration rate. Figure 

20a shows the optimal flight path without any intersection, and Figure 20b presents the 

performance of the ESPO algorithm when the size of the swarm changes. As can be seen 

from the results, regardless of the swarm size, the algorithm converges to the best mini-

mum cost function. By comparing the results of Figures 19b and 20b, EPSO can reduce the 

cost function by 35% compared to PSO. In other words, the EPSO algorithm was able to 

reduce the total length of the route from 79 km to 49 km. 

 

 

(