Monitoring respiratory rate (RR) is essential for early identification of respiratory and metabolic abnormalities. However, the limitations of contact-based sensors and the lack of reliability in many contactless methods make continuous and accurate monitoring difficult in non-clinical settings. To address these challenges, we introduce RespFormer, an edge-optimized, motion-guided temporal-frequency multimodal fusion transformer framework for real-time, contactless RR estimation and breathing pattern classification. RespFormer integrates dense optical flow analysis with temporal, statistical, and frequency-domain features derived from video sequences and enhances them through a multi-stage signal processing pipeline. These features are modeled using an ensemble of three time series transformer architectures (ETSformer, Temporal Fusion Transformer, and Informer) to capture distinct aspects of temporal dynamics. A shared attention-based refinement module enhances the feature representations, and final predictions are fused using a stacking-based meta-learner. We validate RespFormer on a multimodal dataset comprising synchronized RGB, NIR, and IR video data, including a custom in-house dataset captured under various conditions. Experimental results demonstrate that RespFormer achieves a mean absolute error (MAE) ≈ 0.98 bpm, improving prediction accuracy by ≈ 11% and reducing memory usage by ≈ 26%, while maintaining real-time inference (≈ 1.22 seconds) on resource-constrained devices. Furthermore, RespFormer accurately classifies breathing patterns (normal, bradypnea, tachypnea, and apnea) with 95% accuracy, underscoring it’s potential for practical application in telemedicine, clinical screening, and low-resource healthcare settings.

Human-wildlife conflicts in residential/agricultural settings rely on ineffective deterrents like rodenticides or fences. We introduce InvisibleFence, a modular 3D-printed Vision Pod system with off-the-shelf deterrents. The Vision Pod fuses a 2K camera and 240° motion sensing with an edge-optimized pipeline trained on 44,000 wildlife images of eleven classes—achieving 86.7% mAP. Benchmarking YOLO variants (416p–2K) ensures performance. Upon detection, it sends MQTT commands to drive deterrent units—ultrasonic speakers or lighting/spray modules—that emit tones without affecting humans or pets. InvisibleFence creates adaptive zones that reduce false triggers and limit habituation.

We present a parameter evaluation of FMCW mmWave Radar to assess surface dampness and ruggedness and enhance the navigability of autonomous ground robots. We begin by designing and 3D-printing a mount for the mmWave Radar on a Rosmaster X3 platform. We then collect raw mmWave Radar data from various surfaces (grass, soil, puddles, and mulch) across different seasons (summer, winter, and rainy). Our findings demonstrate that the energy strength parameter is a reliable indicator for assessing the surface: dry surfaces (e.g., dry grass, dry mud) exhibit lower energy strength, whereas wet surfaces display higher values. This dampness and ruggedness factor can be leveraged to develop a cost-map navigability score for autonomous ground robots.