Category: IEEE Intelligent Systems

Advertisement.

Recognizing slanted deck scenes is crucial to security monitoring for protecting ships and making them behave smartly resilient. However, there are multitude of diverse structures that are designed as slanted planes due to rough maritime environments. Traditional methods for scene understanding from 3D point clouds or RGB-D data are energy-consuming and memory intensive, which makes those models less reliable in a resource-constrained system of limited compute, memory, and energy resources on ships. In this study, we present an approach to understanding deck scenes, including slanted structures, using a low-cost monocular camera without prior training. New clusters of slanted angle projections are extracted. The vanishing points of slanted non-Manhattan angle projections are estimated. These slanted planes can be reshaped by compositions of non-Manhattan angle projections. Combined with Manhattan planes, a deck scene can be approximated by Manhattan and non-Manhattan planes. Unlike deep learning-based algorithms, this approach requires no prior training or knowledge of the camera’s internal parameters. Experimental results demonstrated that the method can successfully elucidate diverse elements, including slanted structures, meeting safety monitoring requirements using a resource-constrained monocular camera in a deck environment.

Nowadays, artificial intelligence (AI) is bursting in many fields, including critical ones, giving rise to reliable AI that means ensuring safety of autonomous decisions. As the false negatives may have a safety impact (e.g., in a mobility scenario, prediction of no collision, but collision in reality), the aim is to push them as close to zero as possible, thus designing “safety regions” in the feature space with statistical zero error. We show here how sensitivity analysis of an explainable AI model drives such statistical assurance. We test and compare the proposed algorithms on two different datasets (physical fatigue and vehicle platooning) and achieve quite different conclusions in terms of achievable performance that strongly depend on the level of noise in the dataset rather than on the algorithms at hand.

Because encryption is a fundamental security building block, existing encryption techniques like AES, Twofish, Blowfish, and Triple DES are constantly under the threat of being compromised. We introduce a simple graph-theoretic encryption method named CryptoCliqIn using clique injection and prove that the decryption of this encryption without the appropriate key is #P-complete. We have shown that the proposed model does not introduce delays in encryption and decryption times and provides a more secure mechanism compared to some of the existing encryption mechanisms. Finally, an adaptation of CryptoCliqIn in an intelligent system is discussed under the setup of intelligent and smart building.

Advertisement.

Artificial agents that support people in their daily activities (e.g., virtual coaches and personal assistants) are increasingly prevalent. Since many daily activities are social in nature, support agents should understand a user’s social situation to offer comprehensive support. However, there are no systematic approaches for developing support agents that are social situation aware. We identify key requirements for a support agent to be social situation aware and propose steps to realize those requirements. These steps are presented through a conceptual architecture centered on two key ideas: 1) conceptualizing social situation awareness as an instantiation of “general” situation awareness, and 2) using situation taxonomies for such instantiation. This enables support agents to represent a user’s social situation, comprehend its meaning, and assess its impact on the user’s behavior. We discuss empirical results supporting the effectiveness of the proposed approach and illustrate how the architecture can be used in support agents through two use cases.

Prospective authors are requested to submit new, unpublished manuscripts for inclusion in the upcoming event described in this call for papers.

Knowledge-based entity prediction (KEP) is a novel task that aims to improve machine perception in autonomous systems. KEP leverages relational knowledge from heterogeneous sources in predicting potentially unrecognized entities. In this article, we provide a formal definition of KEP as a knowledge completion task. Three potential solutions are then introduced, which employ several machine learning and data mining techniques. Finally, the applicability of KEP is demonstrated on two autonomous systems from different domains; namely, autonomous driving and smart manufacturing. We argue that in complex real-world systems, the use of KEP would significantly improve machine perception while pushing the current technology one step closer to achieving full autonomy.

Prospective authors are requested to submit new, unpublished manuscripts for inclusion in the upcoming event described in this call for papers.

This article presents an online local path planning approach for autonomous drone navigating a 2D plane in an unknown, indoor corridor-like environment. The proposed method utilizes a reinforcement learning approach for training a local path planner for navigation in the said environment. With a continuous actor-critic learning automaton (CACLA) applied for continuous action spaces, the proposed algorithm uses a reward structure that formulates a balancing function that gives reward based on balancing the vehicle between artificial potential hills. The drone thereby learns steering control and obstacle avoidance while maintaining a central aligned position with respect to the unknown hallways or corridors. A novel CACLA algorithm and incorporation of a special experience replay memory for the better converging tendency of drone toward the balancing point have been introduced in this article. The proposed reinforcement learning-based online local path planner has been tested on a simulated drone in Gazebo environment.