In the world of search and rescue operations, robots face a formidable task: maneuvering through a variety of obstacles, and one of the most formidable terrains they encounter is dense vegetation.
(Photo : PHILIP FONG/AFP via Getty Images)
This picture taken on September 22, 2023 shows a demonstration of the X20 robot dog at the DEEP Robotics office in the 2022 Asian Games host city Hangzhou, in China's eastern Zhejiang province.
Offering Distinct Advantages
In their quest to overcome this challenge, robots typically harness a range of sensory tools that include ultrasonic sensors, Lidar (Light Detection and Ranging) technology, infrared sensors, and camera systems to gain awareness of their surroundings.
However, when it comes to navigating through the thick foliage commonly found in outdoor environments, these sensors may fall short in providing a complete solution. Enter a dedicated team of engineers from Carnegie Mellon University, embarking on a mission to crack this intricate puzzle.
David Ologan, a master's student specializing in mechanical engineering at Carnegie Mellon, eloquently describes the heightened complexity of taking robots into the great outdoors, where every action demands meticulous consideration.
"Your system has to be robust enough to handle any unforeseen circumstances or obstructions that you might encounter. It's interesting to tackle that problem that hasn't necessarily been solved yet," Ologan stated on the official press release.
To develop such a robust system, the engineering team is actively devising a reactive walking strategy specifically tailored for a quadrupedal robot.
These four-legged robots offer distinct advantages over their wheeled counterparts when it comes to navigating complex terrains, such as avoiding entanglement with vines and branches, which was a key consideration for the Carnegie team.
Taking Measures Autonomously
The engineers embarked on the formidable task of designing a robot that possesses the remarkable ability to self-monitor its limbs, promptly detecting any potential obstructions. Furthermore, it can autonomously take corrective measures to free its legs from entanglements, a feat that underscores the intricacy of their work.
Ologan emphasized that legged robots possess a unique ability to choose their footholds and step over obstacles rather than simply rolling over them. However, this approach presents a challenge, as each step requires careful consideration of secure footholds.
Interesting Engineering reported that the system's delicacy becomes apparent, as even a slight bump can cause the robot to topple. Therefore, there is a need for methods to effectively respond to external contact.
To address this challenge, the engineering team experimented with various techniques, including high-stepping and a knees-forward approach, to disentangle the robot's legs from obstacles. However, none of these methods worked as effectively as having the robot reactively retract its legs.
This approach proved to be highly beneficial, enhancing the robot's agility and proficiency in navigating environments filled with potential obstacles. What's particularly noteworthy is that this method is adaptable and can be seamlessly integrated into the operating systems of other robots without the need for hardware modifications.
Also Read: AI-Powered Robodogs Master Impressive Feats Thanks to Innovative Algorithm
This adaptability makes the system highly configurable and versatile. The applications for this innovative method are broad and diverse. Effective obstacle avoidance and navigation are critical in various fields of robotics, including autonomous vehicles, drones, industrial robots, and service robots.
As advancements in sensor technology, machine learning, and robotics algorithms continue, Fagen Wasnni Technologies reported that these improvements empower robots to navigate complex and dynamic environments more effectively.
Related Article: This Bio-Inspired Robot Dog Can Run Almost Entirely on Its Own
