teb_local_planner - An Optimal Trajectory Planner for Mobile Robots

This video presents an optimal trajectory planning approach based on the Timed-Elastic-Band approach [1, 2]. The teb_local_planner package is implemented in the Robot Operating System (see below).
The underlying method locally optimizes the robot's trajectory with respect to trajectory execution time, separation from obstacles and compliance with kinodynamic constraints at runtime.
Since local planners such as the Timed-Elastic-Band get often stuck in a locally optimal trajectory as they are unable to transit across obstacles, an extension is implemented [3].
The local planner is able to switch to the current globally optimal trajectory among the candidate set. Distinctive topologies are obtained by utilizing the concept of homology / homotopy classes.
Time-line:
00:00 Title and Introduction
01:00 A deeper Look into the Trajectory Optimization
01:30 Parallel Planning in distinctive Topologies
02:04 Taking polygon and line shaped obstacles into account
02:38 Simulating Robot Navigation
03:56 Collection of Real World Examples
Refer to the ROS Wiki for more details on the software: wiki.ros.org/teb_local_planner
Also watch our new video presenting features introduced in release 0.2: • teb_local_planner - Ca...
Papers describing the approach:
[1] Rösmann, C., W. Feiten, T. Wösch, F. Hoffmann and T. Bertram: Trajectory Modification Considering Dynamic Constraints of Autonomous Robots, 7th German Conference on Robotics, Munich, Germany, May 2012
[2] Rösmann, C., W. Feiten, T. Wösch, F. Hoffmann and T. Bertram: Efficient Trajectory Optimization using a Sparse Model, IEEE European Conference on Mobile Robots, Barcelona, Spain, Sept. 2013
[3] Rösmann, C., F. Hoffmann and T. Bertram: Planning of Multiple Robot Trajectories in Distinctive Topologies, IEEE European Conference on Mobile Robots, Lincoln, UK, Sept. 2015
The Institute of Control Theory and Systems Engineering (RST) at TU Dortmund is engaged in fundamental research in the fields of robotics and computational intelligence. Furthermore the applied research focus is automotive systems and mechatronics. www.rst.e-technik.tu-dortmund.de

[@AugustDerSmarte]

Nice Job. :)

[@giorgoskaramitsos4998]

Really nice work guys!! I would like to ask the robot is able to detect obstacles and moving obstacles through the sensors and the update of the local costmap?

[@erdemerdyniev2600]

which sensors are you using for navigation on this model? only one camera? what about that one on the top looking downward?

[@RstTuDortmund]

Watch a new mobile robotics video from RST!
It describes how our mobile robot ROLF navigates through the environment utilizing our "teb_local_planner" ROS package (wiki.ros.org/teb_local_planner).

[@tejasshastha5457]

Why does it not rotate in place instead of taking a local path off the global plan to reorient?

[@theverner]

Have you mapped the environment beforehand or is it real time?