This Article is written as a summay by Marktechpost Staff based on the paper 'TERP: Reliable Planning in Uneven Outdoor Environments using Deep Reinforcement Learning'. All Credit For This Research Goes To The Researchers of This Project. Check out the paper and post. Please Don't Forget To Join Our ML Subreddit
Autonomous mobile robots are increasingly being used in the real world for activities such as indoor and outdoor surveillance, search and rescue, planetary and space exploration, extensive agricultural surveys, and more. For each of these uses, the robot needs to be able to work on different types of terrain, which can be described by things like color and texture and things like changes in elevation, slope, etc.
The irregularities and slope of the ground mainly determine the stability of a robot, which means that its pitch and roll angles must remain within certain limits. For reliable navigation, robots must recognize dangerous elevation changes and plan most of their paths as well as flat areas. But detecting and navigating in uneven and unstructured environments can be difficult because a complete terrain model with all elevation information is not available. Instead, this information is collected as the robot moves with cameras or LiDAR sensors. Also, the elevation changes cannot be said enough from what you can see in the environment. In the past, the problem has been solved with grid-based data structures like Octomaps and elevation maps, which are 2D grids that show the highest point (in meters) at each grid.
Autonomous mobile robots are already being tested and used for things like package delivery, surveillance, search and rescue missions, planetary and space exploration, and environmental monitoring. For these robots to do their job well, they must be able to work safely and reliably on uneven outdoor terrain without bumping into objects.
Researchers recently developed a new machine learning method that could allow robots to move more easily over uneven outdoor terrain and obstacles. Prior to the researchers’ work, the students began working on their robotic navigation method; they studied how people move in complex outdoor environments. They found it interesting that people ignore the whole environment when moving around. Instead, they focus on parts of the space they deem critical or essential.
The way people walk helped develop a way for robots to find their way around. This method, TERP (Terrain Elevation-based Robot Path planning), is based on a technique they created called deep reinforcement learning (DRL).
The new hybrid machine learning architecture combines the intermediate output results of attention-based DRL networks with a new path planning method. These intermediate results help to find and avoid places in environments that are difficult or dangerous to navigate. The technique uses a fully trained DRL network that uses elevation maps, the robot’s pose, and its goal as inputs to determine an attention mask.
The attention mask created by the team’s algorithm then tells a mobile robot where in its environment it needs to pay particular attention in order to move around without getting lost. This mask is added to the input elevation map at the end of the process to create a 2D navigation cost map. Then this map is used to plot a safe and reliable path for the robot to take to get to a specific location.
In previous work, a significant performance drop was found in end-to-end DRL methods when moving from simulation to real-world terrains. However, with the new hybrid machine learning architecture, the navigation works better.
The attention part of TERP can improve a robot’s sense of space by directing its attention to the areas most critical for the navigation task at hand. The waypoint planner part of their method, on the other hand, makes sure that the robot takes the most profitable path to get to its destination.
The TERP creates relatively stable paths on steep terrain to reduce the risk of the robot tipping over. It can also avoid dangerous areas and obstacles when moving through complex terrain with stationary and moving obstacles.
The researchers used the Husky robot, an unmanned mobile robotic system manufactured by Clearpath Robotics, to test their method in different real-world settings. During their tests, the robots moved in uneven outdoor spaces with a change of up to 4 meters.
The unique hybrid formulation with an attention DRL network for perception and a waypoint planner for navigation has been shown to lead to a high navigation success rate over complex terrain. This means that the method makes it much less likely for a robot to roll over on rough, uneven terrain.
TERP performed very well in the team’s initial tests, suggesting that it can significantly improve the robot’s navigation reliability in harsh outdoor environments. It could be used in the future to improve the functioning of robots in many different situations. For example, it could open up new ways to explore planets and space, do agricultural surveys, and closely monitor the environment.
In addition to terrain height, surface properties such as texture, bumps, and deformability also affect a robot’s ability to navigate complex outdoor situations. Work is underway to have robots learn autonomously in these situations. In addition, other methods are added so that legged robots, such as the Boston Dynamics Spot robot, can move autonomously.