Obstacle avoidance algorithms for mobile robots

In order to achieve autonomous work of mobile robots in everyday living environments and to achieve sophisticated cognitive functions of robot intelligence, they are required to perform complex movements, taking into account the location of the surrounding static objects, the movement of the mobile objects and the achievement of the predetermined goal. This article discusses the most important approaches in this direction, some of which are shown as an example: Common brain-computer interface (BCI) systems that use EEG and EMG signals; motion planning algorithms, situation understanding and decision making algorithms; Learning from Demonstration paradigm for robust motion control achievement; Connectionist (Deep learning) approach for sophistication object grasping and manipulation behaviors and tools using. These approaches achieve the realization of an autonomous car, the ability of robots to solve tasks in the industry, to help people with disabilities or assist humans in their daily lives, to participate in interactive games, to perform actions that require the capture of objects and use tools, prepare food, and perform other intelligent activities.