English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Service robots are appearing more and more in our daily life. The developmentof service robots combines multiple fields of research, from object perceptionto object manipulation. The state-of-the-art continues to improve to make aproper coupling between object perception and manipulation. This coupling isnecessary for service robots not only to perform various tasks in a reasonableamount of time but also to continually adapt to new environments and safelyinteract with non-expert human users. Nowadays, robots are able to recognizevarious objects, and quickly plan a collision-free trajectory to grasp a targetobject in predefined settings. Besides, in most of the cases, there is areliance on large amounts of training data. Therefore, the knowledge of suchrobots is fixed after the training phase, and any changes in the environmentrequire complicated, time-consuming, and expensive robot re-programming byhuman experts. Therefore, these approaches are still too rigid for real-lifeapplications in unstructured environments, where a significant portion of theenvironment is unknown and cannot be directly sensed or controlled. In suchenvironments, no matter how extensive the training data used for batchlearning, a robot will always face new objects. Therefore, apart from batchlearning, the robot should be able to continually learn about new objectcategories and grasp affordances from very few training examples on-site.Moreover, apart from robot self-learning, non-expert users could interactivelyguide the process of experience acquisition by teaching new concepts, or bycorrecting insufficient or erroneous concepts. In this way, the robot willconstantly learn how to help humans in everyday tasks by gaining more and moreexperiences without the need for re-programming.

The State of Lifelong Learning in Service Robots: Current Bottlenecks in Object Perception and Manipulation