Online Control Policy Search for the Robotic All-Terrain Surveyor using Reinforcement Learning
نویسندگان
چکیده
The objective of this project is to devise an online control policy for a particular multi-legged robot. The conceptual robot named robotic all-terrain surveyor (RATS) has 12 legs equally distributed over a spherical surface, approximately the size of a soccer ball. The objective of this system is to have the capability to circumvent complex obstacles through rough terrain. Each leg consists of a pneumatic piston that has enough power to make the whole robot hop. The concept of this robot was made by Boeing Corporation, and it is being designed and developed by the Robotics Institute at Carnegie Mellon University.
منابع مشابه
Design, Analysis, and Learning Control of a Robotic Wind Turbine
Wind power represents one of the most promising sources of renewable energy, and improvements to wind turbine design and control can have a significant impact on energy sustainability. Although wind energy has not been a canonical research area in robotics, we argue here that many robotics techniques (physical system design, modeling and control, and policy learning techniques) can in fact have...
متن کاملLearning Dynamic Manipulation Skills under Unknown Dynamics with Guided Policy Search
Planning and trajectory optimization can readily be used for kinematic control of robotic manipulation. However, planning dynamic motor skills requires a detailed physical simulation, and some aspects of the task, such as contacts, are very difficult to simulate with enough accuracy for dynamic manipulation. Alternatively, manipulation skills can be learned from experience, allowing them to def...
متن کاملRobotic Search & Rescue via Online Multi-task Reinforcement Learning
Reinforcement learning (RL) is a general and well-known method that a robot can use to learn an optimal control policy to solve a particular task. We would like to build a versatile robot that can learn multiple tasks, but using RL for each of them would be prohibitively expensive in terms of both time and wear-and-tear on the robot. To remedy this problem, we use the Policy Gradient Efficient ...
متن کامل1 Supervised Actor - Critic Reinforcement Learning
Editor’s Summary: Chapter ?? introduced policy gradients as a way to improve on stochastic search of the policy space when learning. This chapter presents supervised actor-critic reinforcement learning as another method for improving the effectiveness of learning. With this approach, a supervisor adds structure to a learning problem and supervised learning makes that structure part of an actor-...
متن کاملGuided exploration in gradient based policy search with Gaussian processes
Applying reinforcement learning(RL) algorithms in robotic control proves to be challenging even in simple settings with a small number of states and actions. Value function based RL algorithms require the discretization of the state and action space, a limitation that is not acceptable in robotic control. The necessity to be able to deal with continuous state-action spaces led to the use of dif...
متن کامل