| [eigen] Regression in 3.4.0: EulerAngles computes wrong angles? |
[ Thread Index |
Date Index
| More lists.tuxfamily.org/eigen Archives
]
- To: eigen@xxxxxxxxxxxxxxxxxxx
- Subject: [eigen] Regression in 3.4.0: EulerAngles computes wrong angles?
- From: Juraj Oršulić <juraj.orsulic@xxxxxx>
- Date: Thu, 16 Feb 2023 09:47:13 +0100
- Arc-authentication-results: i=1; mx.microsoft.com 1; spf=pass smtp.mailfrom=fer.hr; dmarc=pass action=none header.from=fer.hr; dkim=pass header.d=fer.hr; arc=none
- Arc-message-signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=microsoft.com; s=arcselector9901; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-AntiSpam-MessageData-ChunkCount:X-MS-Exchange-AntiSpam-MessageData-0:X-MS-Exchange-AntiSpam-MessageData-1; bh=jRyqS9wl1BlMWWuPS6tPHsdEggdEC2eXCCWsYuaI5dI=; b=IXRpNysCMT0u0PJO7E5cZ30MGuvubuhmXM08tdeMq9p+M50+W7hcg/FZz6vVQhBBpffKBjpahfgere5dIov+DSg0HhBUtLqQprlZxQGp4u4WF0EfWy4UDUFTVqs3f67TLKCpqmXKSkfyAfqbJhPBvN8qeghfre7AArs/jF09d/GLBBgh8skS040tjFB1dq2G5A631/wz9/5kprDk1Up9TfCjDFRS6wZPxT15p87NAMDVSdXVFi8NkDX3biFRJ0vLjqcvbqNXSLdnF4h22prtt6wxgqPq99i3qlQf1srLXGCgzs1okA2wSUSeMht2hwXVbHu15K4moukWuSsAbb/7Ew==
- Arc-seal: i=1; a=rsa-sha256; s=arcselector9901; d=microsoft.com; cv=none; b=aJ3fOusJuluMUMFIp7QkQrsbNGa2S+v/PGVzZTfhNN8skW1cmDFlv086SN+OKLePQSFTPnlujOO1Yg/tp3bGD7ivWO/2VmBj5/V0LjIqTNc5TlkkC33NIND1MTgMpgNMqW78Hua+bEBYLZxoNB3sfdNXrRa43wUfygiMTwYcTop3AtxGdhtq2yRZl5JLn1DpweXjzkaDyznGUq0rjPtoiEmchaPB9sQXL7P/8tGf7bp9Iaxu6ccL445Jkvg70pRIEKj0ku4wLUwlNIcqWvpRQKFJ8QbsZPb34QrlEdaCHeoumvQ8bJhIJXmIaKBvVsuv3Pjv0PX0D5AGOGSAqqjixQ==
- Authentication-results: dkim=none (message not signed) header.d=none;dmarc=none action=none header.from=fer.hr;
- Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=ferhr.onmicrosoft.com; s=selector2-ferhr-onmicrosoft-com; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-SenderADCheck; bh=jRyqS9wl1BlMWWuPS6tPHsdEggdEC2eXCCWsYuaI5dI=; b=WqbsREwSJETqyz/VcIKP75lP8+uKTrz078nwnvvVeqbOdyRdVAVM9lqkFs1nOm234eDZDDr7EAMB3V76qijdKxnNw+93GysqMtKpG5B7kAIuBxirqcupZJLhCQgdbDsch1KEQweA8HPjhoVki8RjQ30jJtCE+ppzXUCys30mTSk=
Hello everyone,
this was one of the cases where you think you’re going crazy, until you realize that might be a core library bug. In this case, it looks like that a regression in 3.4.0 causes a totally wrong result given by Eigen::EulerAngles for a certain gymbal lock configuration when the input quaternion has some tiny floating point noise.
I have opened an issue on Gitlab with a code sample which reproduces the issue on Godbolt. When the library version on Godbolt is changed 3.3.9, it computes the correct answer.
Here’s the issue:
Could someone confirm that I’m not going crazy? :D I know that EulerAngles is in Eigen-unsupported, but still it shouldn’t be messing up this bad.
Regards, Juraj
—
Laboratory for Robotics and Intelligent Control Systems
Faculty of Electrical Engineering and Computing
University of Zagreb