Enabling Autonomous Exploration for Robots
Researchers have created a range of robotic systems and planners that empower robots to expedite exploration, venture into uncharted territories, and generate highly precise and detailed maps. These systems enable robots to achieve these tasks autonomously, navigating and mapping without any human intervention.
Carnegie Mellon University’s Autonomous Robotics Team
However, a team of researchers from Carnegie Mellon University’s Robotics Institute is at the forefront of creating the next generation of explorers – robots. In fact, this Autonomous Exploration Research Team has successfully developed a suite of robotic systems and planners that empower robots to embark on faster explorations, probe even the most challenging and unfamiliar environments, and generate highly accurate and detailed maps. The notable feature of these systems lies in their autonomy, allowing the robots to navigate and map without any human involvement.
Ji Zhang, a systems scientist in the Robotics Institute, explains the convenience of the system, stating that it can be deployed in various environments, such as department stores or disaster-stricken residential buildings, and the robots can autonomously build real-time maps while exploring without requiring any human presence.
Unraveling the World with Versatile Robotic Systems
Over the course of three years, the team has worked on exploration systems and has successfully explored and mapped multiple locations, including underground mines, a parking garage, the Cohon University Center, and other indoor and outdoor spaces on the CMU campus. The system’s adaptability allows it to be attached to almost any robotic platform, transforming them into modern-day explorers. In the process of testing, the team utilizes adapted motorized wheelchairs and drones.
The robotic systems operate in three modes: one mode allows human control over the robot’s movements while the autonomous systems prevent collisions with objects; in another mode, a person can select a point on the map, and the robot will navigate to that location; the third mode is pure exploration, where the robot autonomously explores the entire space and creates a map.
Enhancing Efficiency with Advanced Sensor Integration
The team integrated a variety of sensors, such as a 3D scanning lidar sensor, a forward-looking camera, and inertial measurement unit sensors, along with a sophisticated exploration algorithm. This enables the robot to determine its current location, trace its previous path, and decide on the next steps in the exploration process. The resulting systems are much more efficient than previous methods, as they produce more comprehensive maps and reduce the algorithm’s runtime by half.
A Game-Changing Solution for Challenging Settings
Notably, these new systems are well-suited for operating in challenging conditions with low-light and spotty communication, such as caves, tunnels, and abandoned structures. The group’s exploration system was used in Team Explorer, a joint entry from CMU and Oregon State University in DARPA’s Subterranean Challenge, where it ranked fourth in the final competition and won the Most Sectors Explored Award for mapping more of the route than any other team.
The team is committed to open-sourcing all their work, with the aim of empowering society with the capabilities of building autonomous exploration robots. Chao Cao, a Ph.D. student in robotics and the lead operator for Team Explorer, highlights that once this fundamental capability is attained, it opens up a plethora of possibilities for various applications, ranging from delivery services to search-and-rescue missions.
Read the original article on ScienceDaily.
Read more: Field Trials of Autonomous Military Robots Are Being Conducted in Estonia.
