How to remote control your car via ROS-Mobile App This blog shows my current progress on turning a remote controlled car into an autonomous robot. A RaspberryPi 4 runs ROS Noetic and sends pwm signals to the throttle and steering of the rc car via a 16 channel servo driver board. The goal is to…
In May 2022 I have submitted my master’s thesis with the title ‘Navigation Strategies for Passively Articulated Mobile Robots Based on Path Planners Existing in ROS’. You want to peek inside it or even contribute to the research topic?
Here is the Link to Google Drive.
Here is the Link to the project on GitHub
Mobile robots that autonomously execute tasks in outdoor terrain receive more and more attention. There are many well-researched steering congurations such as dierential-drive, skid-steering or car-like robots. But passively articulated robots in contrast are still under-researched, despite of their many advantages. In order to close the research gap in mobile robotics, this thesis focuses on navigation strategies for passively articulated robots. These robots usually consist of a front and rear carriage that are connected with each other via a passively articulated joint. First of all, path planners are examined that already exist in Robot Operating System (ROS). The working principles and algorithms of the two global planners global_planner and sbpl_lattice_planner, as well as of the two local planners dwa_local_planner and teb_local_planner are thoroughly explained. Since the navigation stack in ROS employs one global and one local planner, the resulting four combinations are tested on a simulated and physical robot. However, the generic path planners in ROS can navigate passively articulated robots by instructing forward motions only. In order to support motion reversals, several approaches are developed that externally intervene the navigation process by means of a virtual navigation frame. Luckily, this does not require modications of the planning algorithms themselves. Furthermore, numerous contributions of other researchers on actively articulated vehicles or truck-trailer systems are investigated and analyzed with the intention to nd planning algorithms capable of taking the complete (x_front; y_front; \alpha_front; \alpha_joint) robot state into account. Taking all these insights into consideration, further navigation strategies for future research are presented.