IEEE Transactions on Robot Learning (T-RL) publishes fundamental advances in development and use of Artificial Intelligence (AI) methods that address challenges specific to robotic and automation systems, incorporating the constraints and opportunities present in physical systems, and tackling key barriers to the full deployment of AI in a robot from perception to control.

Examples of topics within the scope of T-RL include, but are not limited to, transfer learning for efficient knowledge transfer across robotic platforms, tasks, and environments; methods tailored to key robotic challenges such as limited pre-existing data, poor generalizability, slow learning timescales and low robustness in physical deployment; algorithmic advances in AI-based control of robots and automation systems that offer safety and reliability guarantees. Topics may include robot learning algorithms for human-robot collaboration, such as methods for prediction of human behavior; explainable and transparent AI-driven control of robots and automation systems; building trust; and attributing responsibility.

T-RL publishes work with both theoretical and practical significance. This includes foundational mathematical and algorithmic advances and application papers, including benchmarks describing robotics and automation use cases that enable reproducibility and comparisons between approaches. Applications in real hardware in addition or to complement evaluation in simulation are expected. Tutorial and review papers provide an overview of developments and/or use of statistical learning algorithms and approaches that advance new capabilities in robotics and automation.

All IEEE journals require an Open Researcher and Contributor ID (ORCID) for all authors. ORCIDs enable accurate attribution and improved discoverability of an author’s published work. The author will need a registered ORCID in order to submit a manuscript or review a proof in this journal.

Follow these steps to link a PaperCept account to a registered ORCID. Note that PINs from different sub-domains can be linked to a single ORCID.

1. Go to https://ras.papercept.net/journals/trl/scripts/login.pl.  Create a PIN if you do not have one.
2. If you do not already have an ORCID, follow the link “Register an ORCID,” create an ORCID, and return to the PaperCept login page.
3. Follow the link “Register your ORCID with any of your PaperCept PINs,” enter your registered email or the ORCID and password, and then click on Authorize.
4. Once you authorize the ORCID to be linked to your PIN you will be returned to the PaperCept subdomain from where you accessed ORCID.
5. Follow “Log in to any account associated with this ORCID” to complete associating your PIN to the ORCID.
6. If you have inadvertently created multiple PINs in the same subdomain, you can associate all PINs with the same ORCID.

If you have not previously submitted a paper to T-RL, you are strongly encouraged to peruse recent issues to familiarize yourself with the style and technical level of typical T-RL articles. You are also strongly encouraged to read Surviving the Review Process by former T-RO EiC Seth Hutchinson.

The peer-reviewing process of the articles submitted to this journal fulfills the requirements outlined in the IEEE Publication Services and Products Board Operations Manual (https://pspb.ieee.org/images/files/files/opsmanual.pdf). Each published article was reviewed by a minimum of two independent reviewers using a double-anonymous peer review process, where the identities of the reviewers are not known to the authors and the identities of the authors are not known to the reviewers (see https://www.ieee-ras.org/publications/rules-for-the-double-anonymous-review-process).

For each paper that proceeds to the full review process, the editorial board of the IEEE Transactions on Robot Learning (T-RL) attempts to obtain three high-quality reviews.  By submitting to T-RL, you and your co-authors agree to provide up to three high-quality reviews of other T-RL submissions.

The IEEE TRANSACTIONS ON ROBOT LEARNING publishes both regular and survey papers. If a paper is intended for a Special Issue with an open call for papers, the submission’s cover letter must clearly state this. As of January 2025, due to the double-anonymous reviewing processes, the category of evolved papers is no longer available.

• Regular papers describe a significant result or set of results that are original, of high impact and significance, and extend the state-of-the-art in robot learning. If your submission is based on previously published papers, your manuscript must not be just a mere extension of the previous version(s) filling in, e.g., details of proofs or additional corollaries, additional experiments, or more detailed background and narrative. Rather, the new submission must contain new results of substantive research significance and impact beyond the previous papers. If your manuscript has evolved from a previously published paper, special care needs to be taken. As with every other work, you need to clearly specify in which way it goes beyond the state of the art. However, when you reference your previous work, avoid phrases such as “in our previous work […], we …” and rather write “a recent work […] proposed…” so as not to reveal your identity. The TRANSACTIONS ON ROBOT LEARNING will NOT consider submissions that are wholly or partially under review for a conference or any other publication. This is considered an ethical violation and such submissions will be returned immediately without review.
  Whenever authors use parts of previously published papers, the sources need to be cited appropriately. Although such papers are no longer required to add a dedicated paragraph in the introduction describing the extension to the previous paper, it is probably good to add a paragraph of the form: “This manuscript substantially extends the work by Your Student T et al. [X] by a novel …” and add a citation of the form
  [X], Your Student T, …
  Also when authors re-use figures or large text fragments, they should make sure that the previous work is properly cited.
  If the copyright holder of the source with the components copied into the submission to the TRANSACTIONS ON ROBOT LEARNING is neither IEEE nor the author(s), the author(s) should be aware of a potential infringement. In such cases, the authors need to negotiate with the publisher of their previous publication to obtain the right to copy the corresponding parts into the publication of their submitted manuscript.
  Accepted regular papers will be invited for presentation at upcoming IEEE-RAS-sponsored or co-sponsored conferences such as ICRA, IROS, or CASE.

• Survey papers should cover emerging research topics in robot learning or have a tutorial or review nature of sufficiently large interest to the Transactions audience. Authors of survey papers must submit a proposal, as a single self-contained PDF file, to the Editor-in-Chief prior to submission. This proposal should address the need for the survey article (including the potential audience, T-RL relevance, and an analysis of any existing survey papers on the general topic), give a summary of the authors’ qualifications to write the survey (a survey article should be written by leading experts in the field, and it is generally preferred that a survey be a joint effort among research groups at different institutions who are not frequent collaborators), and provide a proposed bibliography. (If the paper is already written, you should provide the paper itself in place of the bibliography.)  The EiC, in conjunction with the Senior Editorial Board, will then determine whether the proposed article may be submitted for review.A good survey paper must be more than an organized list of references. It must provide an authoritative and useful perspective and organization of the history and the state of the art of the field. It should include important references but leave out those that are minor or derivative contributions. The survey should not be biased toward the authors’ own work. It should organize the material according to themes as appropriate (for example, applications or principles or technologies employed), and it should provide fair comparisons or critiques where appropriate. The authors should be acknowledged experts in the field. In short, a good survey paper should provide a perspective and distillation of the field that is a significant contribution in itself.

Both regular and survey submissions may be a maximum of 12 pages long in Transactions format (see Manuscript Preparation for more details on the format). Multipart papers will not be accepted.

Papers submitted to a special issue are handled similarly to normal contributions.

Rejected Manuscripts

Manuscripts that have substantial deficiencies, for example, with respect to their presentation or the state of the art should be rejected. Authors of a rejected manuscript should receive a detailed justification of the causes for rejection. Authors of rejected manuscripts are discouraged from submitting a revised version. However, they are not prevented. If authors submit a revised version of a previously rejected manuscript, they are expected to provide a statement of changes justifying the reasons for a resubmission. The EiCs will normally assign the resubmitted manuscript to the SE and AE supervising the previously rejected manuscript unless they are retired in the meantime. The editorial board will carefully analyze resubmissions of previously rejected manuscripts and also consider an editorial rejection decision for such manuscripts.

Revised and Resubmitted Manuscripts

A manuscript should receive a “revise and resubmit decision” if it contains deficiencies that can be corrected within a reasonable amount of time. If the decision on a submission to T-RL is “revise and resubmit” (RR), authors will be allowed to submit a revised version and a statement of changes. The revised manuscript should be formatted so that any changes can be easily identified by the reviewers, by using, e.g., colored or bold text to indicate revised passages. In addition to the largely revised version of a paper, the authors are required to also upload a single pdf file containing a reply to the comments provided by the reviewers and the list of changes made to the manuscript paper. After resubmission, the EiCs will normally assign the manuscript to the SE and AE previously supervising the manuscript, unless they are retired in the meantime. Revised and resubmitted manuscripts are treated as novel submissions. A second “revise and resubmit” decision for such manuscripts should be avoided. Papers not reaching the bar for a “conditionally accept decision” should preferably be rejected.

Conditionally Accepted Manuscripts

A manuscript should receive a “conditionally accepted decision” if it contains minor deficiencies that can be corrected within 60 days. Conditionally accepted special collection papers must be corrected within 30 days. If the decision on a submission to T-RL is “conditionally accepted” [CR/CC], authors will be allowed to submit a revised version and a statement of changes. The revised manuscript should be formatted so that any changes can be easily identified by the reviewers, by using, e.g., colored or bold text to indicate revised passages. In addition to the revised version of your conditionally accepted paper, the authors should also upload a single PDF file containing the authors’ reply to the comments provided by the reviewers and the list of changes made to the manuscript. The EiCs will assign resubmissions of previously conditionally accepted manuscripts to the SE and AE supervising the previous version unless one or both of them have retired in the meantime. The evaluation of such a manuscript should take no longer than one month. The AE is expected to quickly evaluate the updated manuscripts potentially with the assistance of the reviewers assigned to the previous version. A second “conditionally accept decision” should be avoided.

Accepted Manuscripts

Manuscripts fulfilling all requirements for a publication in the IEEE Transactions on Robot Learning should receive an “accept decision”. Authors of accepted manuscripts are requested to prepare a final version of their manuscript strictly the indications provided in the reviews, if any. No other significant unreviewed changes (including to the bibliography and references) are allowed. The submission of the final version is due no later than 30 days from the notification date. When the final version of an accepted paper is received, the SE has the possibility to verify that all required changes have been made in the final version and may request the AE to provide support in this phase. Although rarely, an SE can withhold publication until all required changes are satisfactorily made. In case this can not be achieved by authors within 180 days from submission, the manuscript could be effectively withdrawn from publication.

Editorial (Summary) Rejections

Editorial reject decisions are to be used mainly for incomplete or otherwise clearly unsuitable papers (see below the official IEEE policy for more details). All editorial rejections must be approved by the EiCs. If the reason for editorial rejection is a lack of technical contribution, IEEE rules require that at least three editors concur in such a decision. For these cases, the EiCs and at least two members of the Editorial board (usually the SE and AE) have to consent. If an AE comes to the conclusion that an assigned manuscript should be summarily rejected, the AE informs the corresponding SE with a sufficiently detailed justification. If an SE thinks that a manuscript should be summarily rejected, the SE assigns it to an AE asking for the AE’s opinion. If both the AE and the SE agree that the paper should be summarily rejected, they generate a sufficiently detailed decision letter (typically drafted by the AE and revised by the SE). They then inform the EiC about their proposed decision, provide the EiC with a decision letter, and ask for approval. If the EiC, the SE, and the AE agree, the SE issues the editorial reject decision. The decision should not be executed without approval by the EiC. An editorial reject decision should be issued no later than three weeks after the submission. If the EiC, the SE, and the AE disagree in their assessment of the manuscript, it should be sent out for full review.

Editors may prescreen articles immediately after they are submitted and before they are transmitted to referees for evaluation. The purpose of such prescreening is to verify that the article adheres to minimum criteria set forth by IEEE, as well as by the organizational unit responsible for the specific publication. Typical prescreening measures include the following:

• The author(s) have followed the IEEE guidelines for style.
• The author(s) have not obviously violated IEEE Policies.
• The article is comprehensible (in other words, not so poorly written that it is unreadable).
• The subject and contents of the article meet the scope of the periodical or a specific issue.
• The article meets a minimum criterion for technical substance established for the periodical.

If the Editor assesses that a submission has met prescreening criteria 3.a through 3.d but has not met criterion 3.e, the Editor shall consult with at least two members of the editorial board for concurrence. Rejection on the basis of criterion 3.e shall require the general agreement of the Editor and those consulted. An article is considered in review if it passes the prescreening process and is forwarded to referees. An “administrative reject” refers to an article that does not meet the prescreening measures and is, therefore, returned to the author(s) with an explanation.

T-RL policies on Open Access (OA) publishing are the same as general IEEE policies. The majority of all IEEE transactions, journals, and letters offer a hybrid option, which permits both traditional subscription-based content (which is free for the author, apart from possible fees such as over-length charges) as well as OA access, author-pays content. For all hybrid journals, the quality of the review process for OA and traditional articles is the same. Any OA papers published within a hybrid journal will be included in all media types offered by that title.

Some authors using government research funding or university consortium funding may be required to publish in OA journals. In addition, authors may choose to pay to publish to gain the largest possible audience for their work.

T-RL allows OA for authors who request it.  This option is made available upon acceptance of the paper.  Whether a paper will be OA or not is not known during the review process and does not affect the process.

January 1, 2026 and after the OA article processing charge (APC) is USD $2,800. Other applicable fees (e.g., over-length paper charges) remain unchanged.

• IEEE Members receive a 5% discount.
• IEEE Society Members receive a 20% discount.
• Discounts cannot be combined.
• Discounts do not apply to undergraduate and graduate students.
• Information about IEEE’s Low and Lower-middle Income Country Program is available.

IEEE seeks to maximize the rights of its authors and their employers to post preprint versions of an article on the author’s personal website, on a server operated by the author’s employer, or on a preprint server operated by an approved not-for-profit third party such as ArXiv or TechRxiv. This allows rapid dissemination of the work for comment and establishes a time-stamp associated with the results. Once the paper is accepted, however, IEEE requires that the paper be replaced by the accepted version of the paper, with an acknowledgment that the paper will appear in the IEEE Transactions on Robot Learning and the following copyright notice: “© 20xx IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.” The version posted on the preprint server should also include a link to the paper on IEEE Xplore and the Digital Object Identifier (DOI) when they are available. (Only the accepted version, not the IEEE-edited published version, may be posted to the preprint server.)

If the submission is a revised version of a paper that was previously reviewed by the Transactions, whether the previous version received a decision of conditionally accept, revise and resubmit, or reject, the new submission must be accompanied by a separate file with a clear response to the AE’s summary and the reviews, including indications of how the paper has been modified in response. This file must not exceed ten pages. (Note that authors are discouraged from submitting a revision of a rejected paper.) A good practice is to simply copy all AE and reviewer comments into your response, separated into sections for the AE and each reviewer, and to intersperse your replies where appropriate. This makes explicit which comments you are responding to and which you are not. Each of your replies should (1) include any comments you have in response and (2) briefly point to where the paper has been modified if it has been modified. The revised manuscript should be formatted so that any changes can be easily identified by the reviewers, by using, e.g., colored or bold text to indicate revised passages. If the previous version was rejected, the Cover Note must include a reference to the ID number of the rejected paper. Failure to adhere to these guidelines will result in your paper being returned without review.

More details can be found at the Submission Procedures page.

Authors have the option of suggesting potential reviewers in their cover letter. Please note that whether the suggested reviewers are in fact asked to review the submission will be left entirely to the discretion of the handling associate editor and editor. Suggested reviewers should not have any conflict of interest with any of the authors, or have any past or current relationship with the authors that in any way hinders their ability to provide an objective and unbiased review. Also, if the authors feel that a particular person will not be able to provide an unbiased review of their manuscript and wish to exclude this person as a reviewer, they should provide a detailed explanation of the reasons, e.g., any publications or common work that creates a conflict of interest between the authors and the potential reviewer.

Page Length – Submissions must adhere to the page length limits described under Types of Submissions.  These page limits include everything needed for publication, e.g., figures, tables, references, author photos, and biographies (no more than 100 words each).

Paper Format – All manuscripts should be submitted in the double-column format of the Transactions using an IEEE style file. Instructions and templates (LaTeX or Word) are downloadable from the IEEE Author Center, together with other useful tools for authors.

If you use LaTeX, then papers should be prepared using the command

\documentclass[journal]{IEEEtran}

with a default font size of 10 pt.

When using the IEEE style format, the first (title) page will contain the paper title and each author’s name, affiliation, and full address (mailing address, email address, and fax number), with the corresponding author clearly indicated, the abstract (no more than 200 words), the keywords (index terms), and the beginning of the main text of the paper. Do not use a cover page (the appearance should be just as published articles in the Transactions). T-RL employs the double-anonymous reviewing process and thus the authors’ names should not be indicated in the PDF of the paper. The type of submission (regular or survey paper) may be indicated in the page header. Note that the PDF will be stamped by the system upon submission, so it is better not to interfere with this by overloading footer and header.  Author biographies are required for final accepted papers but are optional for initial submissions.

To enhance the appearance of your article on IEEEXplore®, a graphical abstract can be displayed along with traditional text. The graphical abstract should provide a clear, visual summary of your article’s findings by means of an image, animation, video, or audio clip. Please note that the graphical abstract is considered a part of the technical content of the article, and must be submitted together with your article for peer review. For more information about graphical abstracts and their specifications, please visit the IEEE Author Center.

Figures, Tables, and References – All figures and tables must be numbered and cited in the text. References must be in a separate reference section at the end of the paper, with items referred to by numerals in square brackets. References must be completed in IEEE style as follows:

• Author(s), first initials followed by last name, title in quotation marks, periodical, volume, inclusive page numbers, month, and year.
• Books: Author(s), first initials followed by last name, title, location, publisher, year, chapter, page numbers.

To guard against citation manipulation, IEEE discourages the citation of articles that do not report original research and may not have been peer-reviewed, e.g., editorials, summaries, book reviews, interviews, columns, perspectives, in memoria, etc.

Keywords – Include a list of keywords on the first page of the submission. A list of significant keywords (also named index terms) should be included in the first page of each submitted paper. During one of the submission steps at the T-RL PaperCept site you will have the opportunity to choose keywords.

PDF File Testing – Sometimes, PDF files generated with software or operative systems of Far East countries may have a font embedding problem with Chinese/Japanese/Korean fonts. This prevents the paper to be shown on screen or printed correctly with standard Western system configurations. We cannot expect that all reviewers are equipped with software extensions to work properly with such PDF files, slowing down the review process. Therefore, authors should carefully generate their PDF files and are urged to check them before submission by using the web facility accessible at the T-RL PaperCept site ras.papercept.net/journals/trl. The minimum version of an acceptable PDF is currently 1.4.

Copyright and E-Copyright Form – It is the policy of the IEEE to own the copyright to the technical contributions it publishes on behalf of the interests of the IEEE, its authors, and their employers, and to facilitate the appropriate reuse of this material by others. To comply with U.S. copyright law, authors are required to sign an IEEE copyright transfer form with the final manuscript. It returns to authors and their employers full rights to reuse their material for their own purposes. Authors of papers accepted for the IEEE TRANSACTIONS ON ROBOT LEARNING should fill the electronic copyright transfer form (E-Copyright) at the time of final submission of the material for publication.

Chinese, Japanese, and Korean (CJK) Author Names – IEEE supports the publication of Chinese, Japanese, and Korean (CJK) author names in the native language alongside the English versions of the names in the author list of an article. For more information, please visit the IEEE Author Center.

Inquiries – All pertinent correspondence relating to publications should be sent to the Editor-in-Chief. For routing inquiries, please contact the Editorial Assistant.

T-RL allows multimedia attachments (typically videos or datasets) to accompany a paper.

Multimedia can be “playable” files (.mpeg, .avi, .wav, .mov, .midi, etc.) or “dataset” files (e.g., raw data with programs to manipulate them). Such material is intended to enhance the contents of a paper, both in clarity and in added value. IEEE has set up general guidelines for the submission of multimedia, ranging from the format to the description of content and user requirements, and to the way this material should be referenced to in the body of the paper. This information is available in the “Multimedia Materials” section of the IEEE Author Center.

In preparing this material, it is recommended to follow the general IEEE guidelines right from the initial submission. In addition, specific guidelines hold for the TRANSACTIONS ON ROBOT LEARNING:

• The paper should be self-contained, i.e., fully readable and understandable independently from the multimedia material.
• See the RAS Video Submission Guidelines
• A “ReadMe.txt” file (in ASCII text format) should accompany the multimedia, describing:
  • Minimum requirements for the user. In particular, provide the version of the software that is required to play/run the submitted files. Include the name of the software, the version number, and any special requirements for the player.
  • Contact Information. The author should provide contact information in case users have questions regarding the multimedia material (IEEE will not provide any technical support for this).
• A “Summary.txt” file (in ASCII text format) should also be prepared, describing in 5 sentences or less the contents or value of the multimedia object.

Authors can submit their multimedia material (say, a video, the ReadMe file, and the Summary file) as a single zipped archive file (max 50 MB) during one of the submission steps at the T-RL Papercept site ras.papercept.net/journals/trl.

The multimedia material will be reviewed together with the submitted paper. Once a paper and its associated multimedia material is accepted, the latter will be available in the T-RL page within IEEE Xplore next to the PDF of the paper.

Details of the paper submission process can be found on the Submission Procedures page.

The use of content generated by artificial intelligence (AI) in an article (including but not limited to text, figures, images, and code) shall be disclosed in the acknowledgments section of any article submitted to an IEEE publication. The AI system used shall be identified, and specific sections of the article that use AI-generated content shall be identified and accompanied by a brief explanation regarding the level at which the AI system was used to generate the content. 

The use of AI systems for editing and grammar enhancement is common practice and, as such, is generally outside the intent of the above policy. You can read more about RAS Generative AI Guidelines (https://www.ieee-ras.org/publications/guidelines-for-generative-ai-usage).