A new motion-planning model lets robots decide how to reach an objective by exploring the environment, observing supplementary means, and taking advantage of what they've learned before in similar situations
Researchers at the Massachusetts Institute of Technology have now developed a way to assist robots in navigating surroundings further like humans do. A paper explaining the model was presented the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).
Prevalent motion-planning algorithms generate a tree of likely decisions that branches out until it finds a good path for navigation. One downside, however, is that these algorithms hardly learn. The researchers built a model that combines a planning algorithm with a neural network that learns to recognize paths which could lead to an optimum result, and uses that knowledge to lead the robot's movement in any given surroundings.