Adaptive and intelligent robot task planning for home service: A review. (January 2023)

• Record Type: Journal Article
• Authors: Li, Haizhen; Ding, Xilun
• Is Part Of: Engineering applications of artificial intelligence. Volume 117:Part B(2023)
• Publication Date: 2023-01
• Subjects: Robot task planning -- Home environment -- Uncertainty and dynamic -- Adaptability and intelligence; Engineering -- Data processing -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Ingénierie -- Informatique -- Périodiques; Intelligence artificielle -- Périodiques; Systèmes experts (Informatique) -- Périodiques; Artificial intelligence; Engineering -- Data processing; Expert systems (Computer science); Periodicals; 620.00285
• DOI: 10.1016/j.engappai.2022.105618
• Languages: English
• ISSNs: 0952-1976

Abstract

Abstract: The uncertainty and dynamic of home environment present great challenges to the task planning of service robots. The nature of the home environment is highly unstructured, with a diversity of tasks and a variety of objects. In addition, human, the service principle of robots, dynamically changes the environment and shows a large number of demands in the form of commands. Therefore, interaction is necessary. Robots must have the ability to adapt to complex scenes and act intelligently. Robot task planning for serving in unstructured and dynamic home scenes has been overlooked in the existing reviews. This paper firstly explores primary sources of uncertainty in-depth and summarizes three challenges, i.e., reliable planning under uncertain and incomplete information, efficient planning for complex tasks, and scalable planning for task generalization. Secondly, key solutions to the three challenges are discovered; the advantages and limitations of various task planning techniques are assessed from the perspective of adaptability and intelligence. Finally, the promising avenues toward overcoming uncertainty, improving satisfaction and comfort of users, and enhancing learning skills and generalization are proposed. Highlights: Robot task planning for home service with uncertainty and dynamics from the perspective of adaptability and intelligence is discussed. Three challenges i.e. reliable planning under uncertain and incomplete information, efficient planning for complex tasks and scalable planning for task generalization are summarized. Three types of mainstream solutions to the three challenges are summarized. Advantages and limitations of various task planning techniques are assessed. Future research directions in overcoming the uncertainty, improving the satisfaction and comfort of users and enhancing learning skills and generalization are outlined.