The challenge of navigation in environments with dynamic objects continues to be a central issue in the study of autonomous agents. While predictive methods hold promise, their reliance on precise state information makes them less practical for real-world implementation. This study presents visual forecasting as an innovative alternative. By introducing intuitive visual cues, this approach projects the future trajectories of dynamic objects to improve agent perception and enable anticipatory actions. Our research explores two distinct strategies for conveying predictive information through visual forecasting: (1) sequences of bounding boxes, and (2) augmented paths. To validate the proposed visual forecasting strategies, we initiate evaluations in simulated environments using the Unity engine and then extend these evaluations to real-world scenarios to assess both practicality and effectiveness. The results confirm the viability of visual forecasting as a promising solution for navigation and obstacle avoidance in dynamic environments.

A two-state self-paced brain-computer interface (SP-BCI) divides human mental states into two parts: intentional control (IC) and no-control (NC) states. The state during which the user wants to activate the BCI by is called the IC state. False positive rate (FPR) plays a critical role in evaluating a SP-BCI's usability, where FPR refer to the rate incorrectly classifying NC states. Therefore, development of a method which can not only minimize the FPR and also control it to be smaller than any predefined threshold is of vital importance. In this paper, an imbalanced support vector machine (ISVM)-based FPR control scheme is proposed. This method is able to force the FPR of a two-state SP-BCI to achieve the desired FPR, and this effect is independent of subjects and the feature extraction methods used. In this study, the IC state refers to the state during which the subjects perform an instructed motor imagery task, while during the NC state the subjects are instructed to relax. Spectral power and common spatial pattern (CSP) were used to extract features from EEG signals. Results on the IC and NC EEG data from four participants demonstrate the validity of the proposed ISVM-based FPR control scheme in controlling the FPR to be equal to or smaller than any thresholds, including 5% and 0%.