Preprint Article Version 1 Preserved in Portico This version is not peer-reviewed

Estimation and Planning Differential Drive Robot Simultaneously - Technical Report

Version 1 : Received: 8 February 2019 / Approved: 11 February 2019 / Online: 11 February 2019 (09:00:14 CET)
Version 2 : Received: 6 January 2020 / Approved: 7 January 2020 / Online: 7 January 2020 (05:25:17 CET)

How to cite: Hoang, T. Estimation and Planning Differential Drive Robot Simultaneously - Technical Report. Preprints 2019, 2019020084. Hoang, T. Estimation and Planning Differential Drive Robot Simultaneously - Technical Report. Preprints 2019, 2019020084.


Estimation and planning play a vital role in the construction of an autonomous navigation framework. However, these problems are often considered separately, while planning gives robot a free-collision path towards the desired goal, estimation algorithm presents the executed trajectory in the sense that it has to be closed to the ground truth path as much as possible. Recently, a unified probabilistic framework, which supports solving these problems simultaneously, dubbed STEAP has been proposed. Nevertheless, its current version is only designated for an omni wheels robot, which allows robot to move and turn in vertical direction. Differential drive robot, on the other hand, though limited to move along only one direction, has been used in various situations due to its flexibility and lower cost in hardware designing. Thus, in this extension, our aim is to control a differential drive robot via STEAP. Moreover, in a more complicated environment such as labyrinth or maze, the original STEAP sometimes fails to find a path. Indeed, this problem is mainly caused by the poor initialization and the non-linearity in optimizer constraints. In our implementation, instead of dealing with these constraints, we employ a global planner algorithm such as Dijkstra or RRT to treat STEAP as an effective local planner module that focus on following the global path. Consequently, the experimental results show that the extended STEAP not only able to navigate a differential drive robot but also in a more complicated and unstructured environment.

Supplementary and Associated Material


SLAM; Trajectory Optimization; Motion Planning; Gaussian Process; STEAP; GPMP2; Mobile Robot; Differential Drive Robot


Computer Science and Mathematics, Robotics

Comments (0)

We encourage comments and feedback from a broad range of readers. See criteria for comments and our Diversity statement.

Leave a public comment
Send a private comment to the author(s)
* All users must log in before leaving a comment
Views 0
Downloads 0
Comments 0
Metrics 0

Notify me about updates to this article or when a peer-reviewed version is published.
We use cookies on our website to ensure you get the best experience.
Read more about our cookies here.