Heterogeneous-ants-based path planner for global path planning of mobile robot applications

Mobile robots can be applied to a wide range of problems, and the demand for these applications has risen in recent years, increasing interest in the study of mobile robotics. Many studies have examined the path planning problem, one of the most important issues in mobile robotics. However, the grid paths found by traditional planners are often not the true shortest paths or are not smooth because their potential headings are artificially constrained to multiples of 45 degrees. These paths are unfit for application to mobile robots because the high number of heading changes increases the energy required to move the mobile robot. Some studies have proposed a post-processing step to smooth the grid path. However, in this case, the post-smoothed path may not necessarily find the true shortest path because the post-smoothed path is still constrained to headings of multiples of 45 degrees. This study attempts to develop a global path planner that can directly find an optimal and smoother path without post-processing to smooth the path. We propose a heterogeneous-ants-based path planner (HAB-PP) as a global path planner to overcome the shortcomings mentioned above. The HAB-PP was created by modifying and optimizing the global path planning procedure from the ant colony optimization (ACO) algorithm. The proposed algorithm differs from the traditional ACO path planning algorithm in three respects: modified transition probability function for moving ants, modified pheromone update rule, and heterogeneous ants. The simulation results demonstrate the effectiveness of the HAB-PP.