Path-planning modules for Autonomous Vehicles: Current status and challenges

A detailed survey of the available literature on path planning of Autonomous Ground Vehicle (AGV) is conducted, including the overview of single-robot control architectures, different path-planning approaches, analyses of current sensor systems and different velocity estimation techniques. In order to achieve the full autonomous operation of a mobile robot, path or motion planning, i.e., the planning of a collision-free path from a start to goal position through a collection of obstacles, is the fundamental task in the field of autonomous control systems. As AGVs are used in a wide variety of applications to perform autonomous tasks, organising their intelligence plays a key role in successfully programming a robot for a particular application and applying the right control architecture makes the autonomous control problem easier to solve. For autonomous vehicles, sensors play an important role in acquiring different attributes of the working environment and, by extracting meaningful information from these data, the autonomous system can acquire knowledge about its environment. Moreover, different velocity estimation techniques are reviewed in the context of dealing with the dynamic obstacles. A brief review of the available literature on path-planning approaches and techniques is provided.