Adapted ACO Algorithm for Energy-Efficient Path Finding of Waste Collection Robot

• Published in: International Conference on Control, Automation and Systems (Online) pp. 469 - 474
• Main Authors: Tomitagawa, Koki, Anuntachai, Anuntapat, Chotiphan, Supannada, Wongwirat, Olarn, Kuchii, Shigeru
• Format: Conference Proceeding
• Language: English; Japanese
• Published: ICROS 27.11.2022
• Subjects: Adaptation models; and ant colony optimization (ACO); Ant colony optimization; Energy consumption; Energy efficiency; energy optimization; Heuristic algorithms; path finding; Production facilities; robot; Simulation; Waste collection
• ISSN: 2642-3901
• DOI: 10.23919/ICCAS55662.2022.10003892

Waste collection is a major concern of many companies with large areas of facility, e.g., buildings or factories, where there are many trash bins at various dumping points. Therefore, they require human labor to handle, which is a major cost of consideration. Currently, there are research works using robots for waste collection instead of humans. There is a challenge for waste collection robots in terms of energy consumption to pick up the waste at various dumping points efficiently. The factors related to the energy consumption of waste collection robots are directly related to the distance and waste weight that the robots have to collect and carry from the trash bins at various dump points along the paths. This paper presents the adapted ant colony optimization (ACO) algorithm to find the energy-efficient paths of the waste collection robots. The adapted ACO algorithm uses the waste weight in the trash bin as path heuristic information between two dumping points to determine the state transition probability for finding the most energy-efficient path. The experiment was conducted by the simulation to compare the result with the conventional ACO algorithm that uses distance as the path heuristic information. The simulation results expressed that the adapted ACO algorithm provided the most energy-efficient path under the number of nodes and waste weights specified better than the conventional ACO algorithm.