Control of a Nonholonomic Mobile Robot Using Neural Networks

Introduction

这篇文章主要就是提出了一个control structure that makes possible the integration of a kinematic controller and a neural network computed-torque controller for nonholonomic mobile robots.

Mobile robot navigation can be classified into three basic problems: tracking a reference trajectory, following a path and point stabilization. Some nonlinear feedback controllers have been proposed for solving these problems. The main idea behind these algorithms is to find suitable velocity control inputs which stabilize the closed-loop system. 这篇文章这三种问题都要解决。

现在解决方法中，一般To compute the vehicle control inputs, it is assumed that there is “perfect velocity tracking”. 这样的假设会带来三个问题：

• the perfect velocity tracking assumption does not hold in practice
• disturbances are ignored
• complete knowledge of the dynamics is needed

这篇文章提出的方法其实就是通过一个NN controller来解决上面的问题。
下面来简要介绍一下本文提出的方法——the nonlinear kinematic-NN backstepping controller：
It provides a rigorous method of taking into account the specific vehicle dynamics to convert a steering system command into control inputs for the actual vehicle.
所以这个controller其实就是分为两个部分：

1. feedback velocity control inputs are designed for the kinematic steering system to make the position error asymptotically stable
2. an NN computed-torque controller is designed such that the mobile robot’s velocities converge to the given velocity inputs.

第一部分可以是任意的kinematic-based method, 第二部分is independent of the navigation problem because its function is to compute the torque inputs based on approximating the nonlinear dynamics of the cart.