Flocks of birds, schools of fish, and insect swarms are examples of the coordinated motion of a group that arises spontaneously from the action of many individuals. Here, we study flocking behavior from the viewpoint of multiagent reinforcement learning. In this setting, a learning agent tries to keep contact with the group using as sensory input the velocity of its neighbors. This goal is pursued by each learning individual by exerting a limited control on its own direction of motion. By means of standard reinforcement learning algorithms we show that (i) a learning agent exposed to a group of teachers, i.e., hard-wired flocking agents, learns to follow them, and (ii) in the absence of teachers, a group of independently learning agents evolves towards a state where each agent knows how to flock. In both scenarios, the emergent policy (or navigation strategy) corresponds to the polar velocity alignment mechanism of the well-known Vicsek model. These results (a) show that such a velocity alignment may have naturally evolved as an adaptive behavior that aims at minimizing the rate of neighbor loss, and (b) prove that this alignment does not only favor (local) polar order, but it corresponds to the best policy or strategy to keep group cohesion when the sensory input is limited to the velocity of neighboring agents. In short, to stay together, steer together.
Deep neural networks are rapidly emerging as data analysis tools, often outperforming the conventional techniques used in complex microfluidic systems. One fundamental analysis frequently desired in microfluidic experiments is counting and tracking the droplets. Specifically, droplet tracking in dense emulsions is challenging due to inherently small droplets moving in tightly packed configurations. Sometimes, the individual droplets in these dense clusters are hard to resolve, even for a human observer. Here, two deep learning-based cutting-edge algorithms for object detection [you only look once (YOLO)] and object tracking (DeepSORT) are combined into a single image analysis tool, DropTrack, to track droplets in the microfluidic experiments. DropTrack analyzes input microfluidic experimental videos, extracts droplets' trajectories, and infers other observables of interest, such as droplet numbers. Training an object detector network for droplet recognition with manually annotated images is a labor-intensive task and a persistent bottleneck. In this work, this problem is partly resolved by training many object detector networks (YOLOv5) with several hybrid datasets containing real and synthetic images. We present an analysis of a double emulsion experiment as a case study to measure DropTrack's performance. For our test case, the YOLO network trained by combining 40% real images and 60% synthetic images yields the best accuracy in droplet detection and droplet counting in real experimental videos. Also, this strategy reduces labor-intensive image annotation work by 60%. DropTrack's performance is measured in terms of mean average precision of droplet detection, mean squared error in counting the droplets, and image analysis speed for inferring droplets' trajectories. The fastest configuration of DropTrack can detect and track the droplets at approximately 30 frames per second, well within the standards for a real-time image analysis.
Tracking droplets in microfluidics is a challenging task. The difficulty arises in choosing a tool to analyze general microfluidic videos to infer physical quantities. The state-of-the-art object detector algorithmYou Only Look Once (YOLO)and the object tracking algorithmSimple Online and Realtime Tracking with a Deep Association Metric (DeepSORT)are customizable for droplet identification and tracking. The customization includes training YOLO and DeepSORT networks to identify and track the objects of interest. We trained several YOLOv5 and YOLOv7 models and the DeepSORT network for droplet identification and tracking from microfluidic experimental videos. We compare the performance of the droplet tracking applications with YOLOv5 and YOLOv7 in terms of training time and time to analyze a given video across various hardware configurations. Despite the latest YOLOv7 being 10% faster, the real-time tracking is only achieved by lighter YOLO models on RTX 3070 Ti GPU machine due to additional significant droplet tracking costs arising from the DeepSORT algorithm. This work is a benchmark study for the YOLOv5 and YOLOv7 networks with DeepSORT in terms of the training time and inference time for a custom dataset of microfluidic droplets.
Active fluid droplets are a class of soft materials exhibiting autonomous motion sustained by an energy supply. Such systems have been shown to capture motility regimes typical of biological cells and are ideal candidates as building-block for the fabrication of soft biomimetic materials of interest in pharmacology, tissue engineering and lab on chip devices. While their behavior is well established in unconstrained environments, much less is known about their dynamics under strong confinement. Here, we numerically study the physics of a droplet of active polar fluid migrating within a microchannel hosting a constriction with adhesive properties, and report evidence of a striking variety of dynamic regimes and morphological features, whose properties crucially depend upon droplet speed and elasticity, degree of confinement within the constriction and adhesiveness to the pore. Our results suggest that non-uniform adhesion forces are instrumental in enabling the crossing through narrow orifices, in contrast to larger gaps where a careful balance between speed and elasticity is sufficient to guarantee the transition. These observations may be useful for improving the design of artificial micro-swimmers, of interest in material science and pharmaceutics, and potentially for cell sorting in microfluidic devices.
30 August 2021 in Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences
We present a deep learning-based object detection and object tracking algorithm to study droplet motion in dense microfluidic emulsions. The deep learning procedure is shown to correctly predict the droplets’ shape and track their motion at competitive rates as compared to standard clustering algorithms, even in the presence of significant deformations. The deep learning technique and tool developed in this work could be used for the general study of the dynamics of biological agents in fluid systems, such as moving cells and self-propelled microorganisms in complex biological flows. This article is part of the theme issue ‘Progress in mesoscale methods for fluid dynamics simulation’.
21 August 2021 in The European Physical Journal Plus
The state-of-the-art deep learning-based object recognition YOLO algorithm and object tracking DeepSORT algorithm are combined to analyze digital images from fluid dynamic simulations of multi-core emulsions and soft flowing crystals and to track moving droplets within these complex flows. The YOLO network was trained to recognize the droplets with synthetically prepared data, thereby bypassing the labor-intensive data acquisition process. In both applications, the trained YOLO + DeepSORT procedure performs with high accuracy on the real data from the fluid simulations, with low error levels in the inferred trajectories of the droplets and independently computed ground truth. Moreover, using commonly used desktop GPUs, the developed application is capable of analyzing data at speeds that exceed the typical image acquisition rates of digital cameras (30 fps), opening the interesting prospect of realizing a low-cost and practical tool to study systems with many moving objects, mostly but not exclusively, biological ones. Besides its practical applications, the procedure presented here marks the first step towards the automatic extraction of effective equations of motion of many-body soft flowing systems.