Successful integration of radio navigation & RTK setup

Posted on by

In late February, the DLR-Kommunikation und Navigation and Cognitive Neuroinformatics group (CNI) performed integration tests concerning the radio navigation and RTK setup in preparation of the VaMEx3 field tests. We conducted tests at CNI in Bremen, using two radio navigation modules comprising a USRP, a Nuc Computer, and a radio antenna. These modules transfer ranging measurements between each others and will eventually be integrated onto the Robotnik Summit XL rovers. During the upcoming field tests additional seven payload boxes will be deployed, each equipped with the radio navigation module. Through exchange of ranging information a successful navigation will be performed as well as positioning of the rovers will be robustified.

Successful test of radio communication and ranging.

Moreover, the multi-robot graph simulation in Gazebo Fortress was extended to assess the radio navigation outside of field tests. We designed a preliminary model of the payload boxes, which is able to publish its pose in Gazebo. A ROS2 node receiving the position information via the Gazebo ROS bridge, computes ranges between the payload boxes, and publishes them. These ranging measurements can enhance SLAM and positioning robustness for the multi-robot graph SLAM or other algorithms. We plan to implement the interfaces for the multi-robot operations in the near future.

Rovers and payload boxes in Gazebo simulation performing ranging between each other.

Lastly, we tested the GNSS Real-Time Kinematic (RTK) setup of the DLR-KN. We used the RTK corrections with the ZED-F9P and a Moasic H5 receiver. The receivers correctly utilized the corrections, resulting in a reported accuracy of approximately 1cm.

Testing setup for RTK correction data with a dual band GNSS antenna setup.

Multi-Robot Graph SLAM at ICARA 2024

Posted on by

The paper titled “Multi-Robot Graph SLAM using LIDAR” was presented at the International Conference on Automation, Robotics, and Applications (ICARA 2024) in Athens, Greece. This algorithm adopts a decentralized approach, where each robot independently executes its instance of the SLAM process. Through the exchange of SLAM information, each robot constructs a detailed map of its surrounding environment from its own perspective.

The algorithm is deployed in VaMEx3. Check out the source code for this project on GitHub. Feel free to explore and contribute to this collaborative effort in advancing robotics technology.

Above you can see multi-robot graph SLAM simulation results. Nodes, edges, and SLAM point cloud of robot 1 visualized in rviz on the left. The simulation environment and two robots exploring the map are displayed on the right. Only results of robot 1 are visualized, while all information from robot 2 is integrated via the cooperative SLAM.

Looking ahead, the SLAM system is planned to be enhanced to incorporate additional sensor modalities, such as RADAR, radio navigation, and inertial measurement units, to further improve mapping accuracy and robustness.

Successful demo of the VTB at the DLR synergy meeting (October 2023)

Posted on by
VR interaction with multiple rovers

In October 2023 we were able to present a VR demo of our VTB at the fall synergy meeting of the DLR Explorer initiatives. This was a complete success, as there was a lot of interest and positive feedback. The demo included our Mars surface based on HiRISE data, natural phenomena such as sandstorms, VR locomotion and interaction, as well as the latest versions of the swarm members exploring the environment.

Paper at the ACM SUI 2023

Posted on by
Sydney, where the SUI23 took place

The team is happy to announce that our paper entitled “How Observers Perceive Teleport Visualizations in Virtual Environments” was presented and published at the ACM Symposium on Spatial User Interaction (SUI) on October 13, 2023. The goal of our research was to investigate how the popular VR locomotion method teleportation can be visualized and understood by other participants in a shared multi-user environment, such as our VTB. Our results have shown that a visualization, for example by means of a walking animation of the avatar, can lead to significant advantages with regard to important aspects such as presence and spatial awareness. With these findings and a corresponding implementation of suitable visualizations in the VTB, we will be able to significantly increase its user-friendliness in multi-user mode.

VaMEx working meeting in the run-up to the DLR synergy meeting

Posted on by
VaMEx members discussing the upcoming milestones.

The Vamex team used the fall synergy meeting of the DLR Explorer initiatives to hold another productive working meeting, to present the progress of the individual sub-projects and partners, and to talk in more detail – in person – about the upcoming milestones. In conclusion, it can be said that all sub-projects have made very good progress, some exciting but solvable challenges remain and the team is still highly motivated.

First flight of the demonstrator UAV

Posted on by

On September 28, the first flights with the VaMEx demonstrator were performed on the test site of the Institute of Flight Guidance, a decommissioned airfield between Braunschweig and Hanover.

Changing requirements and faulty hardware made it necessary to replace important components of the hexacopter. For example, the drive components and the flight controller were replaced. The new flight controller now also has an Ethernet interface, which makes communication with the onboard computer more robust.

A total of 3 flights were performed, which were used to make basic adjustments to the flight controller and the flight control system. The last flight was already fully automated.

The flight data collected will be evaluated and used to fine-tune the flight controller and the aircraft. In further flight tests, the maximum take-off mass calculated for the final setup of the demonstrator is to be achieved with the help of a replacement payload and the controller parameters adapted to the increased take-off mass so that the first flights for data recording can then be safely completed with the VaMEx sensor payload.

Integration meeting in Oct 2023

Posted on by

The VaMEx Project recently held a three-day meeting to discuss network interface definitions and conduct initial testing phases among the participating project partners. This step was part of the ongoing collaborative process, focusing on establishing clear communication protocols and validating interface functionality through both dummy and partially implemented systems.

In these sessions, partners worked together to detail and agree upon the specifications that will enable their systems to interact effectively. The progress made includes not only the successful definition of network interfaces but also their trial through preliminary implementations, setting the stage for the forthcoming development work. This effort reflects the project’s steady advancement and the practical approach to achieving set objectives in a complex, multi-partner environment.

VaMEx-RGE working meeting

Posted on by

From 29.08. – 31.08. an internal RGE working meeting took place at the University of Bremen. Participants were members of the working group for Cognitive Neuroinformatics, the working group for High-Performance Visualization and DFKI. At times, Andre and Roland from the Computer Graphics and Virtual Reality working group were also present. Together, we clarified many questions on key areas such as SLAM, trajectory planning, VTB connection, mission control and the network architecture.

We see the first deployment of two ubiquity access points as a special milestone. This enabled us to send the first ROS2 messages via a network between the two access points.

Other highlights included a visit to the “MARS findet Stadt” exhibition at the Kulturkirche St. Stephani and dinner at the Pannekoekschip Admiral Nelson.

Extended virtual testbed: ROS2 and a first rover with simulated sensors

Posted on by

We recently managed to integrate ROS2 communication into the UE5-based virtual testbed. This was achieved using and extending the rclUE plugin. ROS2 greatly improves the ability of efficient, decentral communication between all agents of our eventually developed autonomous swarm.

Moreover, we developed and integrated the first swarm unit, a rover, including simulations of various sensors. Specifically, RGB and depth cameras, a Lidar sensor, and an IMU. The collision and friction with the Mars terrain are simulated, too, using the physics engine of the UE5. Although the simulations are rather simple for now, they will be extended incrementally over time.

In order to test the ROS2 communication, we also developed a small external testing application, with which steering commands can be sent to the simulated rover and which shows the sensor output.  

SPEC Meeting 2

Posted on by

On February 16 and 17 the second SPEC meeting of VaMEx3 took place at the University of the Bundeswehr Munich. In total about 20 people participated in the meeting. Michael Schleiss presented a rough mission concept on Mars. The vision is to cover an area of approx. 14km x 14km on Mars to explore and map as well as to perform in smaller regions of interest such as drilling and sampling. to be carried out.

The key outcome of the SPEC meeting is a rough schedule of the VaMEx3 final presentation. Responsible project participants were also assigned to the major hardware and software components so that everyone should know what to do next. Furthermore, the project partners from Anavs, Aerospace Information Technology – University of Würzburg and TUM presented their swarm participants. In addition, the VaMEx3 Virtual Test Bed (VTB) was presented, which will be used to simulate the Mars mission. The VTB combines Unreal Engine 5 and ROS2 communication. In the near future, each project partner has to define requirements for the VTB in order to implement a simulation as realistic as possible.

Another important point was the presentation of possible test areas for the VaMEx3 demonstration. Volker Klein presented different areas in Germany, whereby these should have good accessibility and infrastructure. The focus is on regions of open pit coal mining due to the similarity to the surface of Mars. By fall 2023, further information will be obtained for the regions Garzweiler, Groitscher Dreieck, Lausitz and Eifel.