<?xml version="1.0" encoding="UTF-8"?>
<!-- generator="wordpress/2.0.4" -->
<rss version="2.0" 
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:wfw="http://wellformedweb.org/CommentAPI/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	>

<channel>
	<title>Robotics Research</title>
	<link>http://www.damith.org/research/blog</link>
	<description>Research n stuff on robotics, SLAM and computer vision</description>
	<pubDate>Thu, 21 May 2009 01:52:26 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.0.4</generator>
	<language>en</language>
			<item>
		<title>EKF vs fullslam()</title>
		<link>http://www.damith.org/research/blog/?p=61</link>
		<comments>http://www.damith.org/research/blog/?p=61#comments</comments>
		<pubDate>Wed, 14 Mar 2007 23:55:00 +0000</pubDate>
		<dc:creator>research</dc:creator>
		
	<category>Comments</category>
		<guid isPermaLink="false">http://www.damith.org/research/blog/?p=61</guid>
		<description><![CDATA[U=diag([0.05^2 (5*pi/180)^2]);
zu=2.68; zv=3.06; zd=1.30;
&#160;&#160;&#160;
]]></description>
			<content:encoded><![CDATA[<p>U=diag([0.05^2 (5*pi/180)^2]);</p>
<p>zu=2.68; zv=3.06; zd=1.30;</p>
<p><a href="http://research.damith.org/images/simulated_eg_ekf_vs_fullslam.png"><img src="http://research.damith.org/images/simulated_eg_ekf_vs_fullslam.png" border="1" alt="simulated_eg_ekf_vs_fullslam.png" title="simulated_eg_ekf_vs_fullslam.png" hspace="1" vspace="1" width="171" height="145" style="width: 171px; height: 145px" /></a>&nbsp;<a href="http://research.damith.org/images/simulated_eg_ekf_vs_fullslam_map.png"><img src="http://research.damith.org/images/simulated_eg_ekf_vs_fullslam_map.png" border="1" alt="simulated_eg_ekf_vs_fullslam_map.png" title="simulated_eg_ekf_vs_fullslam_map.png" hspace="1" vspace="1" width="155" height="142" style="width: 155px; height: 142px" /></a>&nbsp;<a href="http://research.damith.org/images/simulated_eg_ekf_vs_fullslam_map_closeup.png"><img src="http://research.damith.org/images/simulated_eg_ekf_vs_fullslam_map_closeup.png" border="1" alt="simulated_eg_ekf_vs_fullslam_map_closeup.png" title="simulated_eg_ekf_vs_fullslam_map_closeup.png" hspace="1" vspace="1" width="143" height="145" style="width: 143px; height: 145px" /></a>&nbsp;</p>
]]></content:encoded>
			<wfw:commentRSS>http://www.damith.org/research/blog/?feed=rss2&amp;p=61</wfw:commentRSS>
		</item>
		<item>
		<title>Ill conditioned Hessian&#8230;</title>
		<link>http://www.damith.org/research/blog/?p=59</link>
		<comments>http://www.damith.org/research/blog/?p=59#comments</comments>
		<pubDate>Tue, 13 Mar 2007 01:07:39 +0000</pubDate>
		<dc:creator>research</dc:creator>
		
	<category>Comments</category>
		<guid isPermaLink="false">http://www.damith.org/research/blog/?p=59</guid>
		<description><![CDATA[
Missing observation causes 
Warning: Matrix is close to singular or badly scaled.&#160;&#160;&#160;&#160;&#160;&#160;&#160;&#160; Results may be inaccurate. RCOND = 1.062233e-37.
&#160;
&#160;
]]></description>
			<content:encoded><![CDATA[<p><a href="http://research.damith.org/images/missing_observation.png"><img src="http://research.damith.org/images/missing_observation.png" border="1" alt="missing_observation.png" title="missing_observation.png" hspace="1" vspace="1" width="264" height="250" style="width: 264px; height: 250px" /></a></p>
<p>Missing observation causes </p>
<p>Warning: Matrix is close to singular or badly scaled.<br />&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Results may be inaccurate. RCOND = 1.062233e-37.</p>
<p>&nbsp;</p>
<p>&nbsp;</p>
]]></content:encoded>
			<wfw:commentRSS>http://www.damith.org/research/blog/?feed=rss2&amp;p=59</wfw:commentRSS>
		</item>
		<item>
		<title>nonholonomic constrants</title>
		<link>http://www.damith.org/research/blog/?p=57</link>
		<comments>http://www.damith.org/research/blog/?p=57#comments</comments>
		<pubDate>Wed, 07 Mar 2007 05:45:08 +0000</pubDate>
		<dc:creator>research</dc:creator>
		
	<category>Comments</category>
		<guid isPermaLink="false">http://www.damith.org/research/blog/?p=57</guid>
		<description><![CDATA[Kinematic constraints are classically
divided into two classes: constraints
that can be integrated to yield constraints
on the position variables,
called holonomic constraints, and constraints
for which this integration is
not possible, called nonholonomic constraints.
A typical example of a nonholonomic
constraint is a wheel rolling
vertically without slipping on a surface.
The constraint on the allowable velocity
(the point of contact of the wheel
with the [...]]]></description>
			<content:encoded><![CDATA[<p align="left">Kinematic constraints are classically</p>
<p align="left">divided into two classes: constraints</p>
<p align="left">that can be integrated to yield constraints</p>
<p align="left">on the position variables,</p>
<p align="left">called holonomic constraints, and constraints</p>
<p align="left">for which this integration is</p>
<p align="left">not possible, called nonholonomic constraints.</p>
<p align="left">A typical example of a nonholonomic</p>
<p align="left">constraint is a wheel rolling</p>
<p align="left">vertically without slipping on a surface.</p>
<p align="left">The constraint on the allowable velocity</p>
<p align="left">(the point of contact of the wheel</p>
<p align="left">with the surface cannot slip in all</p>
<p align="left">directions) cannot be integrated to</p>
<p align="left">yield a constraint on the position of</p>
<p align="left">the wheel. This nonintegrability is</p>
<p align="left">intuitively clear, as illustrated by the</p>
<p align="left">fact that an automobile can go anywhere</p>
<p align="left">it pleases by suitable maneuvering.</p>
<p align="left">(This example hints at the</p>
<p align="left">intimate relationship between nonholonomic</p>
<p align="left">constraints and controllability.)</p>
<p align="left">Loosely speaking, mechanical</p>
<p align="left">systems with holonomic constraints</p>
<p align="left">can be reduced to lower dimensional</p>
<p align="left">mechanical systems without constraints;</p>
<p align="left">for systems with nonholonomic</p>
<p align="left">constraints, this reduction is not</p>
<p align="left">possible, and as a result some distinguishing</p>
<p align="left">features arise</p>
]]></content:encoded>
			<wfw:commentRSS>http://www.damith.org/research/blog/?feed=rss2&amp;p=57</wfw:commentRSS>
		</item>
		<item>
		<title>vision slam sensitivity to sensor alignment</title>
		<link>http://www.damith.org/research/blog/?p=53</link>
		<comments>http://www.damith.org/research/blog/?p=53#comments</comments>
		<pubDate>Tue, 06 Mar 2007 00:25:23 +0000</pubDate>
		<dc:creator>research</dc:creator>
		
	<category>Comments</category>
		<guid isPermaLink="false">http://www.damith.org/research/blog/?p=53</guid>
		<description><![CDATA[both cubicle simulations and the small circle shows bias if the sensors were assumed slightly out of alignment. 
Noticed when running simulation results on fullslam3() and rippleSLAM() the B/2 correction usually used caused large bias resulting in inconsistency.
]]></description>
			<content:encoded><![CDATA[<p>both cubicle simulations and the small circle shows bias if the sensors were assumed slightly out of alignment. </p>
<p>Noticed when running simulation results on fullslam3() and rippleSLAM() the B/2 correction usually used caused large bias resulting in inconsistency.</p>
]]></content:encoded>
			<wfw:commentRSS>http://www.damith.org/research/blog/?feed=rss2&amp;p=53</wfw:commentRSS>
		</item>
		<item>
		<title>Papers ideas of interest</title>
		<link>http://www.damith.org/research/blog/?p=52</link>
		<comments>http://www.damith.org/research/blog/?p=52#comments</comments>
		<pubDate>Thu, 01 Mar 2007 03:47:15 +0000</pubDate>
		<dc:creator>research</dc:creator>
		
	<category>Comments</category>
		<guid isPermaLink="false">http://www.damith.org/research/blog/?p=52</guid>
		<description><![CDATA[Using Treemap as a Generic Least Square - Backend for 6-DOF SLAM: 
Udo Frese -&#62; talks of large maps in 3D in less than a sec using form of LS.


]]></description>
			<content:encoded><![CDATA[<p>Using Treemap as a Generic Least Square - Backend for 6-DOF SLAM: <font></font><font><br />
<p align="left">Udo Frese -&gt; talks of large maps in 3D in less than a sec using form of LS.</p>
<p></font>
</p>
]]></content:encoded>
			<wfw:commentRSS>http://www.damith.org/research/blog/?feed=rss2&amp;p=52</wfw:commentRSS>
		</item>
		<item>
		<title>Things</title>
		<link>http://www.damith.org/research/blog/?p=28</link>
		<comments>http://www.damith.org/research/blog/?p=28#comments</comments>
		<pubDate>Tue, 23 Jan 2007 23:33:58 +0000</pubDate>
		<dc:creator>research</dc:creator>
		
	<category>Comments</category>
		<guid isPermaLink="false">http://www.damith.org/research/blog/?p=28</guid>
		<description><![CDATA[
fullslam to use uvd and preprocessed data.
Sanitizing a data set.
&#160;Error=Estimate-Truth
What&#39;s an exact solution?
Non holonomic constraint

]]></description>
			<content:encoded><![CDATA[<ul>
<li>fullslam to use uvd and preprocessed data.</li>
<li>Sanitizing a data set.</li>
<li>&nbsp;<font color="#ff3300">Error=Estimate-Truth</font></li>
<li>What&#39;s an exact solution?</li>
<li>Non holonomic constraint</li>
</ul>
]]></content:encoded>
			<wfw:commentRSS>http://www.damith.org/research/blog/?feed=rss2&amp;p=28</wfw:commentRSS>
		</item>
	</channel>
</rss>
