Path-Tracking Control of a Non-Holonomic Car-Like Robot with Reinforcement Learning
نویسندگان
چکیده
The problem investigated in this paper is that of driving a car-like robot along a race track and the use of reinforcement learning to find a good control function. The reinforcement learner uses a case-based function approximator to extend the reinforcement learning paradigm to handle continuous states. The learned controller performs similar to the best control functions in both simulation and also in practical driving. 1 CITR, Tamaki Campus, University of Auckland, Private Bag 92019 Computer Science Depatrment of The University of Auckland CITR at Tamaki Campus (http://www.citr.auckland.ac.nz) CITR-TR-47 June 1999 Path-Tracking Control of a Non-Holonomic Car-Like Robot with Reinforcement Learning Jacky Baltes 1, and Yuming Lin 1 Abstract The problem investigated in this paper is that of driving a car-like robot along a race track and the use of reinforcement learning to find a good control function. The reinforcement learner uses a case-based function approximator to extend the reinforcement learning paradigm to handle continuous states. The learned controller performs similar to the best control functions in both simulation and also in practical driving.The problem investigated in this paper is that of driving a car-like robot along a race track and the use of reinforcement learning to find a good control function. The reinforcement learner uses a case-based function approximator to extend the reinforcement learning paradigm to handle continuous states. The learned controller performs similar to the best control functions in both simulation and also in practical driving. 1 CITR, Tamaki Campus, University of Auckland, Private Bag 92019
منابع مشابه
Adaptive fuzzy sliding mode and indirect radial-basis-function neural network controller for trajectory tracking control of a car-like robot
The ever-growing use of various vehicles for transportation, on the one hand, and the statistics ofsoaring road accidents resulting from human error, on the other hand, reminds us of the necessity toconduct more extensive research on the design, manufacturing and control of driver-less intelligentvehicles. For the automatic control of an autonomous vehicle, we need its dynamic...
متن کاملSolving Path Following Problem for Car-Like Robot in the Presence of Sliding Effect via LMI Formulation
One of the main problems of car-like robot is robust path following in the presence of sliding effect. To tackle this problem, a robust mix H2/H∞ static state feedback control method is selected. This method is the well-known linear robust controller which is robust against external disturbance as well as model uncertainty. In this paper, the path following problem is formulated as linear matri...
متن کاملPath Planning with Uncertainty for Car-Like Robots
| this paper presents the rst path planner taking into account both non-holonomic and uncertainty constraints. The case of a car-like robot subject to cumulative and unbounded connguration uncertainty related to bounded control and sensing errors is considered. Assuming the existence of landmarks allowing the robot to relocalize itself in particular places, we present an algorithm that computes...
متن کاملمسیریابی حرکت روباتهای ماشینواره با روش پیشروی سریع
The Robot Motion Planning (RMP) problem deals with finding a collision-free start-to-goal path for a robot navigating among workspace obstacles. Such a problem is also encountered in path planning of intelligent vehicles and Automatic Guided Vehicles (AGVs). In terms of kinematic constraints, the RMP problem can be categorized into two groups of Holonomic and Nonholonomic problems. In the first...
متن کامل