Developing a local planner for autonomous navigation of a Triceratops robot in indoor environments.
In previous work, we developed a visual navigation system for the Triceratops robot, but how can we improve the local planning performance for more robust navigation in complex indoor environments?
As the visual navigation system can not act like LiDAR-based naviagtion to use costmap for local planning, we develop a local planner that can take the laserscan data from depth information by using depth_to_laserscan package.
The local planner is integrated with the existing visual navigation system, allowing for seamless navigation in dynamic obstacle.
Depth to Laserscan Conversion
Obstacle Avoidance with Depth to Laserscan