Navigation of an Autonomous Wheeled Robot in Unknown Environments Based on Evolutionary Fuzzy Control

Abstract

Navigation of a wheeled robot in unknown environments is proposed in this paper. The approach may be applied to navigating an autonomous vehicle in unknown environments, such as parking lots. The navigation consists of three parts: obstacle avoidance behavior, target seeking behavior, and a behavior supervisor. The obstacle avoidance behavior is achieved by controlling the robot to move along an obstacle boundary through evolutionary fuzzy control. In the evolutionary fuzzy control approach, a Pareto set of fuzzy controllers (FCs) is found though a multi-objective continuous ant colony optimization algorithm. Target seeking behavior is achieved by controlling the robot through hybrid proportional-integral-derivative (PID) controllers. The behavior supervisor determines the switching between obstacle avoidance and target seeking behaviors, where the dead-cycle problem is considered. Simulations and experiments were performed to verify the effectiveness of the proposed navigation scheme.

Using a Type-2 Neural Fuzzy Controller for Navigation Control of Evolutionary Robots

“…We compared the proposed method with other evolutionary algorithms. (10)(11)(12)14) The proposed IWOA yields the largest fitness value among the algorithms. The comparison results are presented in detail in Table 2.…”

“…Worst Average STD ABC (10) 0.901 0.891 0.900 0.0043 7 501 DE (11) 0.904 0.894 0.895 0.0069 8 168 PSO (12) 0.915 0.892 0.900 0.0051 9 277 Chou and Juang (14) 0 (14) and (e) IWOA. by a mobile robot, which were obtained using various algorithms in the training environment.…”

“…The quantitative evaluation includes the average distance between the robot and the wall (D avg ), the moving distance of the robot around a (14) and (e) IWOA. (10) 0.3842 56.8100 385 DE (11) 0.3859 56.4471 239 PSO (12) 0.4050 57.5852 378 Chou and Juang (14) 0.5323 58.9955 297 IWOA 0.3933 56.2204 188 circle (D circle ), and the moving time of the robot around a circle (T circle ). Table 3 presents the performance of various algorithms in unknown testing environments.…”

“…(13) Recently, some scholars have applied evolutionary algorithms to develop a controller. Chou and Juang (14) and Lin et al (15) realized mobile robot controllers by using evolutionary algorithms. These controllers (14,15) adopted reinforcement learning and do not need to collect training data in advance.…”

“…Chou and Juang (14) and Lin et al (15) realized mobile robot controllers by using evolutionary algorithms. These controllers (14,15) adopted reinforcement learning and do not need to collect training data in advance. In this study, we focus on WOA, which is based on the hunting strategies of humpback whales, namely, searching for prey, encircling strategy, and the mass net searching strategy.…”