Navigational Analysis of Legged Robot using the Modified African Vulture Optimization Algorithm

The main challenges in the legged robot are navigation, obstacle avoidance, and path development. This study uses a swarm intelligence based meta-heuristic technique the Modified African Vulture Optimization Algorithm (MAVOA), to address the optimum path planning problem. The developed algorithm imitates the foraging behaviour of vultures. The MAVOA dynamically adjusts the movement of vultures to equate exploration and exploitation. This technique is selected to get the optimum path by avoiding obstacles while navigating a legged robot in an unfamiliar, congested environment. When it comes to the legged robot navigation, the path planning problem is solved by selecting four governing constraints: front-side hurdle length (FHL), left-side hurdle length (LHL), right-side hurdle length (RHL), and the direction of turn (DT) that are considered as the inputs and outputs respectively. The main constraints of earlier researches on existing techniques are impotence to account for motion in new environments, challenges in navigating crowded and complex areas, and the need to optimize intricate pathways. The WEBOT software is chosen for simulating a navigation of a legged robot, and result of simulation is differentiated to experimental outputs configured in a lab setting. Both outcomes show that the legged robot successfully navigated to its objectives while avoiding impediments. After the findings were assessed, it was found that there was less than 5% difference between the output of the simulation and the experiment. The approach can be applied to multiple-purpose or discrete optimization issues and future complicated engineering and science disciplines.