Approximating Kinematics for Tracked Mobile Robots

Idea

在运动平面上的瞬时旋转中心是相对于robot是dynamics-dependent,but lie within a bounded area. Thus, optimizing constant ICR positions for a particular terrain results in an approximate kinematic model for tracked mobile robots.

Introduction

Tracked Mobile Robot 现在有如下这几种streer的方式:

  • articulated steering
  • curved track steering
  • skid steering
    这种是现在最常使用的,因为it is simpler from the mechanical standpoint and it achieves a faster response.

skid steering现在存在的问题:
variation of the relative velocity of the two tracks results in slippage as well as soil shearing and compacting in order to achieve steering. 这导致的最终问题就是:kinematics is not straightforward, since it is not possible to predict the exact motion of the vehicle only from the control inputs.

Approach

这篇文章提出的方法就是model kinematics for tracked vehicles by obtaining a geometric analogy with a wheeled differential drive model. 然后对模型进行参数辨识。
做出的假设:planar uniform hard terrain type and mobile robot operation at moderate speeds

Kinematics for Tracked Vehicles

kinematics:
( v x , v y , w z ) = f d ( V l , V r ) (v_{x}, v_{y}, w_{z}) = f_{d}(V_{l}, V_{r}) (vx,vy,wz)=fd(Vl,Vr)
inverse kinematics:
( V l , V r ) = f i ( v x , v y , w z ) (V_{l}, V_{r}) = f_{i}(v_{x}, v_{y}, w_{z}) (Vl,Vr)=fi(vx,vy,wz)

接下来就引入这篇文章中最重要的概念:瞬时旋转中心 Instantaneous Center of Rotation ( I C R v ICR_{v} ICRv)
ICR is defined as the point in the horizontal plane where the motion of the vehicle can be represented by a rotation and no translation occurs.

不仅要entire body的运动,也要考虑the motion of both its tracks on their contact surface with the ground.

左右两边履带的旋转中心在local frame中的定义为 L C R l = ( x L C R l , y L C R l ) LCR_{l} = (x_{LCRl}, y_{LCRl}) LCRl=(xLCRl,yLCRl) and L C R r = ( x L C R r , y L C R r ) LCR_{r} = (x_{LCRr}, y_{LCRr}) LCRr=(xLCRr,yLCRr)

因为履带相比于body多了一个履带自身旋转的自由度,履带上的点的运动是body运动和履带旋转运动的结合,因此履带的LCR和body的LCR不一样。

Note: This definition refers to the LCR of track treads on the ground plane, not to their roll axis.

根据参考文献:
在这里插入图片描述
三个旋转中心都在平行于local X轴的直线。
x I C R v = − v y w z x_{ICRv} = \frac{-v_{y}}{w_{z}} xICRv=wzvy x I C R l = V l − v y w z x_{ICRl} = \frac{V_{l} - v_{y}}{w_{z}} xICRl=wzVlvy x I C R r = V r − v y w z x_{ICRr} = \frac{V_{r} - v_{y}}{w_{z}} xICRr=wzVrvy y I C R v = y I C R l = y I C R r = v x w z y_{ICRv} = y_{ICRl} = y_{ICRr} = \frac{v_{x}}{w_{z}} yICRv=yICRl=yICRr=wzvx

下面分析一个重要性质:
如果robot几乎不旋转的话, x I C R v x_{ICRv} xICRv就会无限大。但是 x I C R l , x I C R r , y I C R v x_{ICRl},x_{ICRr}, y_{ICRv} xICRlxICRr,yICRv are bounded, 因为即使在近似直线运动时,分子和分母都是同阶的无穷小。

由上面几个公式,我们可以推导出forward kinematics和backward kinematics:

  • forward kinematics
    v x = V r − V l x I C R r − x I C R l y I C R v v_{x} = \frac{V_{r} - V_{l}}{x_{ICRr} - x_{ICRl}} y_{ICRv} vx=xICRrxICRlVrVlyICRv
  • inverse kinematics
    V l = v y + x I C R l w z V_{l} = v_{y} + x_{ICRl}w_{z} Vl=vy+xICRlwz
下面我们要分析一下tracked vehicles和wheeled differential drive vehicles的异同

在这里插入图片描述

  • 相同
    对于上图中的那种情况(目前理解是wheeled vehicles没有轮子打滑的情况下),forward和inverse kinematics是相同的
  • 不同
    在没有侧向滑动的情况下,ideal wheels的ICRs are constant and coincident with the ground contact points.
    Track ICRs are dynamics-dependent and always lie outside the track centerlines because of slippage.
    Thus, less track slippage results in track ICRs that are closer to the vehicle.

此外,履带ICRs也会受到robot重心的影响。

下面,我们需要定义一些指标来定量描述一下LCRs对robot控制性能的影响。

  • steering efficiency index
  • normalized eccentricity index

以上讨论的目的是通过 I C R l ICR_{l} ICRl I C R r ICR_{r} ICRr将vehicle外部和内部的动力学影响引入到简单的运动学模型中。
I C R l ICR_{l} ICRl I C R r ICR_{r} ICRr are bounded variables for tracked vehicles and coincide with wheel contact points for ideal differential drive.

Identification of Tracked Kinematic Parameters

The identification process consists on finding three parameters:

  • the local Y coordinate of track ICRs ( y I C R v = y I C R l = y I C R r y_{ICRv} = y_{ICRl}= y_{ICRr} yICRv=yICRl=yICRr)
  • the local X coordinates of both track ICRs ( x I C R r x_{ICRr} xICRr and x I C R l x_{ICRl} xICRl)

以下介绍两种方法:

  • Derivation of Track ICRs from the Simulation of the Dynamic Model
  • 1
    点赞
  • 1
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值