Lifelong Machine Learning-Based Efficient Robotic Object Perception

Dear Colleagues,

This Special Issue (SI) is intended to present scholarly papers that address efficient robotic object perception problem from the perspective of lifelong machine learning. Lifelong machine learning aims to utilizes knowledge from past tasks to efficiently and effectively learn new tasks over a lifetime, which is more suitable for the robotic learning scenarios, i.e., perceive the objects or environments in a never-ending manner. Then, the question emerges: “how to lifelong perceive objects with a robot?” To answer this question, we invite scientists, researchers, and robotic specialists together with academics to share their insights of the lifelong robot perception learning. What will the robot learn in a lifelong manner? What kind of knowledge or experience is most suitable for robot perception? How does the robot learn when encountering a new task? Meanwhile, humans can learn from just one or a handful of examples (i.e., few- or zero-shot learning) with vision–audio–touch senses; can robot achieve very long-term learning in this manner as humans do? All of these are important discussions at the moment and this Special Issue will help all those interested in the topic to promote their vision and ideas.

Overall, this Special Issue focuses on learning with fewer labels for robot multi-modality perception tasks such object classification, object detection, semantic segmentation, robot navigation, SLAM, and many others, and the topics of interest include (but are not limited to) the following areas:

• Life-long/continual/incremental learning methods
• New methods for few-/zero-shot learning
• Robot embodied intelligence
• Federated robot perception learning
• Meta-learning methods
• Self-supervised learning methods
• Novel domain adaptation methods
• Lifelong semi-supervised learning methods
• Lifelong unsupervised learning methods
• Lifelong multi-modal learning methods

Therefore, we invite contributions from experimental researchers and theorists of high-quality manuscripts for publication in this SI.

Dr. Gan Sun
Dr. Hang Zhong
Dr. Qianqian Wang
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Electronics is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• lifelong/continual/incremental learning methods
• new methods for few-/zero-shot learning
• robot embodied intelligence
• federated robot perception learning
• meta-learning methods
• self-supervised learning methods
• novel domain adaptation methods
• lifelong semi-supervised learning methods
• lifelong unsupervised learning methods
• lifelong multi-modal learning methods