In June 2023, the ICRA 2023 RoboMaster Univerisity Sim2Real Challenge (RoboMaster Univerisity Sim2Real Challenge, RMUS), co-organized by the Intelligent Industry Research Institute (AIR) of Tsinghua University and the RoboMaster Organizing Committee, came to a perfect conclusion in Beijing, China.
RMUS Challenge
The RMUS Challenge aims to improve the performance of robots in perception, navigation, and manipulation tasks in the real world through simulation. The core form of the competition is to exchange ores with fully-automated and officially modified RoboMaster EP robots to obtain corresponding scores.
Participating robots use the Sim2Real mode in the 5-minute competition time to compare the accuracy and speed of searching for ore exchange according to the information displayed on the exchange label, and finally rank according to the total score.
▲ Components of RMUS EP
In this competition, all RoboMaster EPs are equipped with the RPLIDAR A2 radar of SLAMTEC, helping the robot to find the target quickly and accurately in the competition
Positioning and mapping and navigation obstacle avoidance
FTM (Champion Team) defend
It is not difficult to find through observation that there are many obstacles in the competition field, such as: L-shaped curves, black shelters, slopes, etc. RPLIDAR A2 uses its own advantages to quickly identify the surrounding environment when facing such obstacles. Tell the “algorithm”, cooperate efficiently, complete obstacle avoidance and navigation, and get more ore.
Excellent ranging accuracy
What I have to mention here is that RPLIDAR A2 performs well when facing low reflectivity objects such as black. It can accurately detect various obstacles in the scene and judge the distance to them. Combined with the algorithm, the robot can locate, navigate and avoid obstacles during walking, find obstacles in time and walk around them.
At the same time, at the competition site, the size of the ore is a cube with a side length of 50mm, the weight is about 98g, and the material is ABS. RPLIDAR A2 ultra high precision performance under close-range objects helps the robot to perfectly complete the ore mining task.
Detect target ores in the scene
The robot detects the target ore in the game scene
LIDAR obtains information such as the orientation, size, and shape of objects in front by continuously emitting and receiving electromagnetic waves. SLAMTEC laser radar can quickly detect the target or information that appears during the robot’s walking through its excellent small object detection ability in the competition.
RPLIDAR X ROS
In the RMUS challenge, ROS was chosen as the standard communication interface, and the AI Habitat platform was connected with the resources of existing robots through ROS-X-Habitat. SLAETEC full range of LIDAR products are compatible with ROS.
RPLIDAR_ros is the first LIDAR in China to launch a corresponding package for ROS developers around the world. The super high cost performance has greatly promoted the use of RPLIDAR in ros, and played a big role in the development of mobile navigation and obstacle avoidance related technologies and talent reserves for robots. big propulsion.
At present, all LIDAR products of SLAMTEC are compatible with the ROS system, and it is also the only domestic LIDAR sensor product officially certified by ROS. At the same time, Silan has always been committed to school-enterprise cooperation with a grateful heart, to drive more people to jointly promote the application and promotion of artificial intelligence technology, and to explore innovative technologies with young forces.
