Deepfake speech detection presents a growing challenge as generative audio technologies continue to advance. We propose a hybrid training framework that advances detection performance through novel augmentation strategies. First, we introduce a dual-stage masking approach that operates both at the spectrogram level (MaskedSpec) and within the latent feature space (MaskedFeature), providing complementary regularization that improves tolerance to localized distortions and enhances generalization learning. Second, we introduce compression-aware strategy during self-supervised to increase variability in low-resource scenarios while preserving the integrity of learned representations, thereby improving the suitability of pretrained features for deepfake detection. The framework integrates a learnable self-supervised feature extractor with a ResNet classification head in a unified training pipeline, enabling joint adaptation of acoustic representations and discriminative patterns. On the ASVspoof5 Challenge (Track~1), the system achieves state-of-the-art results with an Equal Error Rate (EER) of 4.08% under closed conditions, further reduced to 2.71% through fusion of models with diverse pretrained feature extractors. when trained on ASVspoof2019, our system obtaining leading performance on the ASVspoof2019 evaluation set (0.18% EER) and the ASVspoof2021 DF task (2.92% EER).

In this paper, we address the challenge of exploring unknown indoor environments using autonomous aerial robots with Size Weight and Power (SWaP) constraints. The SWaP constraints induce limits on mission time requiring efficiency in exploration. We present a novel exploration framework that uses Deep Learning (DL) to predict the most likely indoor map given the previous observations, and Deep Reinforcement Learning (DRL) for exploration, designed to run on modern SWaP constraints neural processors. The DL-based map predictor provides a prediction of the occupancy of the unseen environment while the DRL-based planner determines the best navigation goals that can be safely reached to provide the most information. The two modules are tightly coupled and run onboard allowing the vehicle to safely map an unknown environment. Extensive experimental and simulation results show that our approach surpasses state-of-the-art methods by 50-60% in efficiency, which we measure by the fraction of the explored space as a function of the trajectory length.

We suggest the first system that runs real-time semantic segmentation via deep learning on the weak microcomputer Raspberry Pi Zero v2 (whose price was $15) attached to a toy drone. In particular, since the Raspberry Pi weighs less than 16 grams, and its size is half of a credit card, we could easily attach it to the common commercial DJI Tello toy-drone ($\times 92.5\times 41$mm). The result is an autonomous drone (no laptop nor human in the loop) that can detect and classify objects in real-time from a video stream of an onboard monocular RGB camera (no GPS or LIDAR sensors). The companion videos demonstrate how this Tello drone scans the lab for people (e.g. for the use of firefighters or security forces) and for an empty parking slot outside the lab. Existing deep learning solutions are either much too slow for real-time computation on such IoT devices, or provide results of impractical quality. Our main challenge was to design a system that takes the best of all worlds among numerous combinations of networks, deep learning platforms/frameworks, compression techniques, and compression ratios. To this end, we provide an efficient searching algorithm that aims to find the optimal combination which results in the best tradeoff between the network running time and its accuracy/performance.

The challenge of mapping indoor environments is addressed. Typical heuristic algorithms for solving the motion planning problem are frontier-based methods, that are especially effective when the environment is completely unknown. However, in cases where prior statistical data on the environment's architectonic features is available, such algorithms can be far from optimal. Furthermore, their calculation time may increase substantially as more areas are exposed. In this paper we propose two means by which to overcome these shortcomings. One is the use of deep reinforcement learning to train the motion planner. The second is the inclusion of a pre-trained generative deep neural network, acting as a map predictor. Each one helps to improve the decision making through use of the learned structural statistics of the environment, and both, being realized as neural networks, ensure a constant calculation time. We show that combining the two methods can shorten the duration of the mapping process by up to 4 times, compared to frontier-based motion planning.

This paper addresses the issue of motion planning in dynamic environments using Velocity Obstacles. Specifically, we propose an adaptive time horizon to truncate the velocity obstacle so that its boundary closely, yet conservatively, approximates the boundary of the set of states from which collision is unavoidable. We wish to develop a representation such that any velocity vector that does not penetrate the velocity obstacle is safe, i.e. an avoidance maneuver exists, and any that does is not. Such clear partitioning between safe and unsafe velocities would allow safe planning with only one step look ahead, and can produce faster trajectories than the conservative trajectories produced when using an infinite time horizon. The computation of the adaptive time horizon is formulated as a minimum time problem, which is solved numerically for each static or moving obstacle. It is used in an on-line planner that generates locally time optimal trajectories to the goal. The planner is demonstrated for static and moving obstacles, and for on-line motion planning in a crowded dynamic environment.

This paper presents a new on-line planner for dynamic environments that is based on the concept of velocity obstacles (VO). It addresses the issue of motion safety, i.e. avoiding states of inevitable collision, by selecting a proper time horizon for the velocity obstacle. The proper choice of the time horizon ensures that the boundary of the velocity obstacle coincides with the boundary of the set of inevitable collision states. This time horizon is determined by the minimum time it would take the robot to avoid collision, either by stopping or by passing the respective obstacle. The planner generates a near-time optimal trajectory to the goal by selecting at each time step the velocity that minimizes the time-to-go and is out of the velocity obstacle. The planner takes into account the shape, velocity, and path curvature of the obstacle's trajectory. It is demonstrated for on-line motion planning in very crowded static and dynamic environments.