A second half-day workshop on “Active methods in autonomous navigation.”
Date: Friday, June 2, 2023, (PM)
Location: ICC Capital Suite 9 | ExCeL London | Royal Victoria Dock | 1 Western Gateway, London E16 1XL
Message from the organisers
Following the previous workshop on “Active methods in autonomous navigation,” which took place at the 2022 Mediterranean Conference on Control and Automation in Athens, Greece, and acknowledging its success, a new series on the same topic follows. More specifically, the participation of influential speakers, the attraction of a large number of researchers, and the drawing of impressive conclusions lead us to the organisation of a second event, which will appeal to a larger crowd, namely the one of the 2023 IEEE International Conference on Robotics and Automation (ICRA), aiming to influence the research community regarding the on-going works.
Abstract
In applications involving the deployment of robots with locomotion capabilities, e.g., humanoids, drones, and automated guided vehicles (AGVs), it is a fundamental requirement that the robot can perceive its environment and decide about it being navigable and the affordances it provides for navigation in general. Nowadays, almost every moving robot is equipped with sensors to observe its working space and gather rich information. However, when robots are starved of computational and sensing resources, simple passive perception approaches, such as building a 3D map before one can act, fall apart. To this end, the active vision, which involves building algorithmic engines around the perception-action synergy loop, offers a promising solution. Active vision endows the robot to actively aim the sensor towards several viewpoints according to a specific scanning strategy. Thus, a vital issue in the active vision systems is that the agent has to decide “where to look” following a plan; that is, the vision sensor must be purposefully configured and placed at several positions to observe a target by taking actions in robotic perception. The intentional acts in purposive perception planning introduce active and purposeful behaviors. In particular, four modes of activeness have been formally identified: by moving the agent itself, by employing an active sensor, by moving a part of the agent’s body, and by hallucinating active movements.
Alignment with the ICRA 2023 “Embracing the future – Making robots for humans.”
According to recent accounts in the field, an “active perceiver knows why it wishes to sense, and then chooses what to perceive, and determines how, when and where to achieve that perception”:
- the “what” question has to do with scene selection and fixation.
- the “when” question has to do with temporal selection, an instant in time, and an extent (a history or episodic memory).
- the “how” question has to do with mechanical alignment, sensor alignment, and priming, and
- the “where” question has to do with viewpoint selection and the relationships between agent pose, sensor pose, and object pose.
Incorporating aspects of the questions above give rise to modern approaches in active perception, coupled with learning for the acquisition of task knowledge and also coupled with the design of the appropriate memories.
Endorsement
The workshop proposal has been endorsed by the technical committees of
Contact
Please get in touch with the organisers with any questions at: activevision.workshop@gmail.com.
