Path planning in autonomous driving systems remains a critical challenge, requiring algorithms capable of generating safe, efficient, and reliable routes. Existing state-of-the-art methods, including graph-based and sampling-based approaches, often produce sharp, suboptimal paths and struggle in complex search spaces, while trajectory-based algorithms suffer from high computational costs. Recently, meta-heuristic optimization algorithms have shown effective performance but often lack learning ability due to their inherent randomness. This paper introduces a unified benchmarking framework, named Reda’s Path Planning Benchmark 2024 (RP2B-24), alongside two novel reinforcement learning (RL)-based path-planning algorithms: Q-Spline Multi-Operator Differential Evolution (QSMODE), utilizing Q-learning (Q-tables), and Deep Q-Spline Multi-Operator Differential Evolution (DQSMODE), based on Deep Q-networks (DQN). Both algorithms are integrated under a single framework and enhanced with cubic spline interpolation to improve path smoothness and adaptability. The proposed RP2B-24 library comprises 50 distinct benchmark problems, offering a comprehensive and generalizable testing ground for diverse path-planning algorithms. Unlike traditional approaches, RL in QSMODE/DQSMODE is not merely a parameter adjustment method but is fully utilized to generate paths based on the accumulated search experience to enhance path quality. QSMODE/DQSMODE introduces a unique self-training update mechanism for the Q-table and DQN based on candidate paths within the algorithm’s population, complemented by a secondary update method that increases population diversity through random action selection. An adaptive RL switching probability dynamically alternates between these Q-table update modes. DQSMODE and QSMODE demonstrated superior performance, outperforming 22 state-of-the-art algorithms, including the IMODEII. The algorithms ranked first and second in the Friedman test and SNE-SR ranking test, achieving scores of 99.2877 (DQSMODE) and 93.0463 (QSMODE), with statistically significant results in the Wilcoxon test. The practical applicability of the algorithm was validated on a ROS-based system using a four-wheel differential drive robot, which successfully followed the planned paths in two driving scenarios, demonstrating the algorithm’s feasibility and effectiveness for real-world scenarios. The source code for the proposed benchmark and algorithm is publicly available for further research and experimentation at: https://github.com/MohamedRedaMu/RP2B24-Benchmark and https://github.com/MohamedRedaMu/QSMODEAlgorithm.

Speed control of DC motors is essential for automated vehicles and four-wheel differential drive (4WD) cars, which are distinct by their high level of maneuverability. The PID controller is one of the most popular techniques for controlling speed, but tuning its parameters is challenging. This paper presents a novel hybrid algorithm, the Moth-Flame Particle Swarm Optimization (MFPSO), which combines moth-flame optimization (MFO) and particle swarm optimization (PSO) to address the slow convergence of MFO and the premature convergence of PSO. The MFPSO is deployed for real-time interactive tuning of the PID controller to control the speed of DC motors in a 4WD car. Additionally, a novel practical procedure is proposed to build a robust four-wheel differential drive and maintain the synchronization of the four DC motors. Simulation results and statistical analysis demonstrate the superior performance of the MFPSO compared with the PSO, MFO, and other hybrid variants (HMFPSO and HyMFPSO), with MFPSO ranking first in the Friedman test on CEC2020/2021 and engineering optimization benchmark problems. Practical results and the transient response analysis of the speed control revealed that MFPSO significantly outperformed the traditional Ziegler-Nichols (ZN) method, MFO, PSO, HMFPSO, and HyMFPSO algorithms. Specifically, the MFPSO algorithm reduced settling time by 34.83%, 21.20%, 20.75%, 22.97%, and 31.59%, and overshoot by 86.11%, 64.99%, 71.02%, 74.37%, and 60.58% compared to the ZN, MFO, PSO, HMFPSO, and HyMFPSO algorithms, respectively. The source code of the proposed algorithm is available at https://github.com/MohamedRedaMu/MFPSO-Algorithm.

This paper aims to find the best variant of Cuckoo Search Algorithm that makes the step size of Lévy flight adaptive. For this reason, we introduce a new variant of CSA called Damped Cuckoo Search (DCS) in which the step size of Lévy flight is adaptive via the concept of damped oscillations that exist in most second order control systems. Moreover, we propose two other methods that tune the step size of Lévy flight based on chaotic maps. Then a deep comparative study is conducted among almost all variants of CSA that appeared in the last decade that modifies the step size of Lévy flight. All these variants are tested on CEC2017 benchmark functions. Statistical analyses are performed using the Friedman test followed by four post-hoc procedures to hold paired comparisons between the proposed DCS and the other CSA variants. Also, graphical statistical analyses are conducted on all variants via Box Plots. Finally, convergence graphs for all the variants are illustrated as well to show the speed of solution improvement over generations. Simulation results prove that the proposed DCS outperforms all other variants with a large degree of significance. Moreover, DCS increases the speed of convergence in comparison with the other variants. The box plot graphs prove that DCS has the most compact distribution for all results obtained in all runs on most functions.