Adaptive Cruise Control tuned by Genetic Algorithm for Safe Distance of Automated Vehicle

Abstract

The importance of safety in transportation has forced developments in automated vehicles. Many control methods have been introduced to ensure safe distance between automated vehicle and another car. This article presents an adaptive cruise control whose parameters are tuned by genetic algorithm for the case study of safe distance in automated vehicle. Two genetic algorithms with two different objective functions, i.e., integral square error (ISE) and integral absolute error (IAE), were evaluated. The results show that genetic algorithms could work well using both functions indicated by the obtained best fitness values through generations. The optimal controller parameter obtained by them also seems to be similar each other, including the distance, velocity and acceleration response of the automated vehicle. It is only the standard deviation that differs them. After 20 running tests, IAE produced 0.0288, 0.0439, and 0.0134 in standard deviation for verr_gain, xerr_gain, vx_gain, respectively, while ISE got 0.0169, 0.0755, and 0.0101. From the total values of standard deviation, we can conclude that GA with IAE performs slightly better since it has smaller value indicating that it has better repeatability.

The Adaptive Cruise Control for Curved Roads Using Archived Crow Search Algorithm

“…When compared to conventional PID control, AI-based control outperforms as presented in the study [11]. Other evidence to show the promising results of using AI-based technique for the case of ACC have also been conducted by the study [12,13], utilizing metaheuristic optimization techniques genetic algorithm (GA) and particle swarm optimization (PSO), respectively. Unfortunately, those reported works are still limited to straight-road case studies, while in fact, there is a possibility to vehicles experiencing a curved road.…”

“…The PI controller has a general equation as written in Eq. (12). 𝑢(𝑡) = 𝑘 𝑝 𝑒(𝑡) + 𝑘 𝑖 ∫ 𝑒(𝜏)𝑑𝜏 𝑡 0 (12) where, u(t) is the control input, e(t) is the error signal, kp is the proportional gain, and ki is the integral gain.…”

“…(12). 𝑢(𝑡) = 𝑘 𝑝 𝑒(𝑡) + 𝑘 𝑖 ∫ 𝑒(𝜏)𝑑𝜏 𝑡 0 (12) where, u(t) is the control input, e(t) is the error signal, kp is the proportional gain, and ki is the integral gain. These parameters play a crucial role in optimizing the steering angle.…”

“…None of the above literatures dealing with the curved road control problem utilized any metaheuristic optimization techniques. As previously mentioned, and highlighted in the [12,13] about the potential use of metaheuristic optimization for straight-road case of ACC control, it can be a new research direction to develop a metaheuristic optimization technique but for the case of curved road control problem. It emerges as a non-trivial problem since ACC for the curved road will not only need to adjust the vehicles' speed, but also to synchronize with the steering control [7].…”