Highly optimized Q‐learning‐based bees approach for mobile robot path planning in static and dynamic environments

Abstract

This paper proposes a new novel approach to find an optimal path for a mobile robot in a two‐dimensional environment. Finding the optimal path is done using the Bees Algorithm (BA) with the Q‐Learning Algorithm. A new method to build the initial population is proposed to find the initial population regardless of the number and location of obstacles in the environment. Q‐Learning is implemented as a local search function of the BA. The hybridization of the BA and the Q‐Learning aims to find the optimal path with a fewer number of iterations of the BA. This method takes advantage of the BA to solve the problem without constraints and the sterilization in the Q‐Learning to find the shortest path. The experiment is run on some different maps to validate the proposed method in the static and dynamic case. The experimental results show the robustness and effectiveness of the proposed method in finding the optimal path. The comparison is executed to view the superiority of this method in finding the shortest path in the comparison of the results of other algorithms.

An Efficient Maritime Route Planning Method Based on an Improved A* with an Adaptive Heuristic Function and Parallel Computing Structure

“…Similarly, the current path planning based on Q-Learning [31,32] is still subject to many obstacles in the application, such as the need for a huge amount of simulation and feedback to generate the path. When considering the synergy of multi-agents, Q-Learning can achieve a comprehensive and better effect.…”

Improved Q-Learning Applied to Dynamic Obstacle Avoidance and Path Planning

“…This article uses Qlearning [9] to find the robot's action to take optimization measures. Bonny et al [10] presented a new method to find the best path for mobile robots in a two-dimensional environment. Use the bees algorithm (BA) and Q-learning algorithm to find the best path.…”

“…In this paper, the sampling method arranges the error values δ j in descending order and then calculates the sampling probability, as shown in (10). If we perform nonuniform sampling, we should adjust the learning rate according to the sampling probability α.…”

Adapting to climate change: the role of metaheuristic algorithms in optimal decision making

“…From enhancing smart microgrids' efficiency and sustainability [9][10][11][12][13][14][15] to pioneering methods for sustainable electrical power generation from biogas [16][17][18], and improving the reliability and performance of battery systems in hybrid electric vehicles [19][20][21][22][23][24], metaheuristic algorithms are reshaping the landscape of technological and environmental solutions. Their role extends to optimizing renewable energy systems, enhancing climate control, and advancing robotic systems through precise control and estimation capabilities [25][26][27][28][29][30][31][32][33][34][35][36][37][38]. Furthermore, the development of *malshabi@sharjah.ac.ae hybrid estimation-based techniques and robust control strategies for various applications, including partial discharge localization [39][40][41] and flexible link manipulators [42][43][44][45][46], demonstrates the algorithms' broad applicability and robustness.…”