Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Studying the effect of multisource Darwinian particle swarm optimization in search and rescue missions
Robotic Swarm Intelligence is considered one of the hottest topics within the robotics research eld nowadays, for its major contributions to di erent elds of life from hobbyists, makers and expanding to military applications. It has also proven to be more effective and effcient than other...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Published: |
AUC Knowledge Fountain
2018
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| Summary: | Robotic Swarm Intelligence is considered one of the hottest topics within the robotics research eld nowadays, for its major contributions to di erent elds of life from hobbyists, makers and expanding to military applications. It has also proven to be more effective and effcient than other robotic approaches targeting the same problem. Within this research, we targeted to test the hypothesis that using more than a single starting/ seeding point for a swarm to explore an unknown environment will yield better solutions, routes and cover more area of the search space within context of Search and Rescue applications domain. We tested such hypothesis via extending existing Particle swarm optimization techniques for search and rescue operations (i.e. Robotic Darwinian Particle Swarm Optimization and we split the swarm into smaller groups that start exploration from di erent seed positions, then took the convergence time average for di erent runs of simulations and recorded the results for quanti cation. The results presented in this work con rms the hypothesis we started with, and gives insight to how the number of robots contributing in the experiments a ect the quality of the results. This work also shows a direct correlation between the swarm size and the search space. |
|---|