Empowering Robots for Autonomous Exploration

2023-07-20

A research team at Carnegie Mellon University's Robotics Institute is creating the next generation of explorers - robots. The Autonomous Exploration Research Team has developed a robot system and planner that enables robots to explore and detect the darkest corners of unknown environments more quickly, and create more accurate and detailed maps. This system allows robots to autonomously perform all these tasks, finding their own way and mapping without human intervention. "You can put it in any environment, such as a post-disaster department store or residential building, and it will be able to navigate," said a scientist from the Robotics Institute. "It will construct maps in real-time and identify the next destination while exploring. You can see everything on the map. Let the robot explore and map the environment without stepping into it." The team has been researching exploration systems for over three years. They have explored and mapped several underground mines, a parking lot, the Cohon University Center, and several indoor and outdoor locations on the CMU campus. The system's computer and sensors can be connected to almost any robot platform, transforming it into a modern explorer. The robot can operate in three modes using the team's system. In one mode, a person can control the robot's movement and direction, while the autonomous system prevents the robot from colliding with walls, ceilings, or other objects. In another mode, people can select a point on the map, and the robot will navigate to that point. The third mode is pure exploration, where the robot sets off on its own, exploring the entire space and mapping it. "This is a very versatile system that can be used for various applications, from delivery to search and rescue," said Howie Choset, a professor at the Robotics Institute. The team combines 3D scanning lidar sensors, forward-facing cameras, and inertial measurement unit sensors with exploration algorithms, allowing the robot to know its location, where it has been, and where it should go next. The resulting system is more efficient than previous methods, creating more complete maps while reducing algorithm runtime by half. The new system can operate in conditions with low light, danger, and unstable communication, such as caves, tunnels, and abandoned buildings. In DARPA's Subterranean Challenge, the CMU and Oregon State University's Team Explorer used a version of this team's exploration system. The team ranked fourth in the final competition but won the "Most Sectors Explored Award" for mapping more routes than any other team. "All our work is open-source. We have no secrets. We hope to benefit society by creating autonomous exploration robots," said Chao Cao, a robotics Ph.D. student and chief operator of the Explorer team. Source: https://techxplore.com/news/2023-07-enabling-autonomous-exploration-robots.html

RECENT AI NEWS

RECENT AI TOOLS