An intelligent fast controller for autonomous wheeled robot path navigation in challenging environments

Abstract

Purpose This paper aims to incorporate one intelligent particle swarm optimization (IPSO) controller to realize an optimum path in unknown environments. In this paper, the fitness function of IPSO is designed with intelligent design parameters, solving the path navigation problem of an autonomous wheeled robot towards the target point by avoiding obstacles in any unknown environment. Design/methodology/approach This controller depends on randomly oriented positions with all other position information and a fitness function. Evaluating the position’s best values, this study gets the local best values, and finally, the global best value is updated as the current value after comparing the local best values. Findings The path navigation of the proposed controller has been compared with particle swarm optimization algorithm, BAT algorithm, flower pollination algorithm, invasive weed algorithm and genetic algorithm in multiple challenging environments. The proposed controller shows the percent deviation in path length near 14.54% and the percent deviation in travel time near 4% after the simulation. IPSO is applied to optimize said parameters for path navigation of the wheeled robot in different simulation environments. Originality/value A hardware model with a 32-bit ARM board interfaced with a global positioning system (GPS) module, an ultrasonic module and ZigBee wireless communication module is designed to implement IPSO. In real-time, the IPSO controller shows the percent deviation in path length near 9%.

Time-Optimal Trajectory Planning for Woodworking Manipulators Using an Improved PSO Algorithm

“…Recently, some advanced technologies, such as machine vision, image processing, artificial intelligence algorithms, etc., have been integrated into automated manipulator and robot systems, which has provided unprecedented opportunities for traditional agriculture and forestry industries to achieve an automate decision-making processes and improve efficiency [46,47]. In this process, the GoldS-PSO algorithm proposed in this paper can be used for the training of neural networks and optimization of control model structure in the intelligent control system [48,49], as well as for the optimization of target recognition, positioning, and tracking technology in visual detection systems [50,51]. Meanwhile, some relevant studies indicate that it has great potential in optimizing forestry management structure [52].…”