Local planning methods for autonomous navigation on sidewalks: a comparative survey

This paper presents a comparative survey of autonomous navigation systems for mobile robots on sidewalks. The kinds of systems mentioned above endow robots with the capability of estimating an optimal trajectory that once followed, allows the robot to move autonomously from a current pose to a target pose on a sidewalk. In addition, they allow avoiding both static and dynamic obstacles reliably and efficiently. An autonomous navigation system functions on data from multiple sensors and a global representation of the environment in which it is located, and allows the robot to perform its motion task satisfactorily without leaving the sidewalk on which it is moving.It is important to mention that an intense search of the state of the art was made around the existing autonomous navigation algorithms. Finally, different tools are mentioned such as ROS2 middleware, the Gazebo simulation program and the Rviz2 visualization program.