On-board Processing Power Research Articles

Fed4UL: A Cloud–Edge–End Collaborative Federated Learning Framework for Addressing the Non-IID Data Issue in UAV Logistics

Artificial intelligence and the Internet of Things (IoT) have brought great convenience to people’s everyday lives. With the emergence of edge computing, IoT devices such as unmanned aerial vehicles (UAVs) can process data instantly at the point of generation, which significantly decreases the requirement for on-board processing power and minimises the data transfer time to enable real-time applications. Meanwhile, with federated learning (FL), UAVs can enhance their intelligent decision-making capabilities by learning from other UAVs without directly accessing their data. This facilitates rapid model iteration and improvement while safeguarding data privacy. However, in many UAV applications such as UAV logistics, different UAVs may perform different tasks and cover different areas, which can result in heterogeneous data and add to the problem of non-independent and identically distributed (Non-IID) data for model training. To address such a problem, we introduce a novel cloud–edge–end collaborative FL framework, which organises and combines local clients through clustering and aggregation. By employing the cosine similarity, we identified and integrated the most appropriate local model into the global model, which can effectively address the issue of Non-IID data in UAV logistics. The experimental results showed that our approach outperformed traditional FL algorithms on two real-world datasets, CIFAR-10 and MNIST.

A Low-Cost Multi-Purpose IoT Sensor for Biologging and Soundscape Activities.

The rapid expansion in miniaturization, usability, energy efficiency, and affordability of Internet of Things (IoT) sensors, integrated with innovations in smart capability, is greatly increasing opportunities in ground-level monitoring of ecosystems at a specific scale using sensor grids. Surrounding sound is a powerful data source for investigating urban and non-urban ecosystem health, and researchers commonly use robust but expensive passive sensors as monitoring equipment to capture it. This paper comprehensively describes the hardware behind our low-cost, small multipurpose prototype, capable of monitoring different environments (e.g., remote locations) with onboard processing power. The device consists of a printed circuit board, microprocessor, local memory, environmental sensor, microphones, optical sensors and LoRa (Long Range) communication systems. The device was successfully used in different use cases, from monitoring mosquitoes enhanced with optical sensors to ocean activities using a hydrophone.

Flight evaluation of simultaneous actuator/sensor fault reconstruction on a quadrotor minidrone

In view of the increase in the number of Unmanned Aerial Vehicles (UAVs) in the commercial and private sectors, it is imperative to make sure that such systems are safe, and thus resilient to faults and failures. This paper considers the numerical design and practical implementation of a linear parameter-varying (LPV) sliding mode observer for Fault Detection and Diagnosis (FDD) of a quadrotor minidrone. Starting from a nonlinear model of the minidrone, an LPV model is extracted for design, and the observer synthesis procedure, using Linear Matrix Inequalities (LMI), is detailed. Simulations of the observer FDD show good performance. The observer is then implemented on a Parrot® Rolling Spider minidrone and a series of flight tests is performed to assess the FDD capabilities in real time using its on-board processing power. The flight tests confirm the performance obtained in simulation, and show that the sliding mode observer is able to provide reliable fault reconstruction for quadrotor minidrone systems.

AudioMoth: A low-cost acoustic device for monitoring biodiversity and the environment

Abstract Environmental sound is a powerful data source for investigating ecosystem health. To capture it, scientists commonly use ruggedized, but expensive acoustic monitoring equipment. In this paper we fully describe the hardware build of a low-cost, small, full-spectrum alternative, called AudioMoth. The credit-card sized device consists of a printed circuit board, micro-controller and a micro-electro-mechanical systems microphone. This simple to construct device facilitates: (1) deployments in remote locations, with a small size and a simple mechanism that allows it to be retrofitted into numerous low-cost ruggedized enclosures; (2) long-term monitoring, with low-power operation; (3) modular expansion, with easy to access general purpose input and output pins; and (4) acoustic detection, with onboard processing power.

Comparative assessment of techniques for initial pose estimation using monocular vision

This work addresses the comparative assessment of initial pose estimation techniques for monocular navigation to enable formation-flying and on-orbit servicing missions. Monocular navigation relies on finding an initial pose, i.e., a coarse estimate of the attitude and position of the space resident object with respect to the camera, based on a minimum number of features from a three dimensional computer model and a single two dimensional image. The initial pose is estimated without the use of fiducial markers, without any range measurements or any apriori relative motion information. Prior work has been done to compare different pose estimators for terrestrial applications, but there is a lack of functional and performance characterization of such algorithms in the context of missions involving rendezvous operations in the space environment. Use of state-of-the-art pose estimation algorithms designed for terrestrial applications is challenging in space due to factors such as limited on-board processing power, low carrier to noise ratio, and high image contrasts. This paper focuses on performance characterization of three initial pose estimation algorithms in the context of such missions and suggests improvements.

TOWARDS MULTIMODAL OMNIDIRECTIONAL OBSTACLE DETECTION FOR AUTONOMOUS UNMANNED AERIAL VEHICLES

Abstract. Limiting factors for increasing autonomy and complexity of truly autonomous systems (without external sensing and control) are onboard sensing and onboard processing power. In this paper, we propose a hardware setup and processing pipeline that allows a fully autonomous UAV to perceive obstacles in (almost) all directions in its surroundings. Different sensor modalities are applied in order take into account the different characteristics of obstacles that can commonly be found in typical UAV applications. We provide a complete overview on the implemented system and present experimental results as a proof of concept.

Autonomous and Adaptive Underwater Plume Detection and Tracking with AUVs: Concepts, Methods, and Available Technology

An autonomous underwater vehicle (AUV) equipped with environmental sensors and an on-board autonomy system can greatly increase the efficiency of environmental data collection and the synopticity of the data set collected simply by autonomously adapting its motion to changes it senses in its local environment. One application of this is tracking ocean features in an unknown ocean environment. This can be accomplished with one or multiple AUVs collaborating in near-real-time using acoustic communications. To further explore one example of this application, this paper focuses on using multiple AUVs to track underwater plumes. We evaluate various types of plumes (e.g., hydrothermal vent plumes, algal blooms, oil leaks), how each plume type may be detected and its spatial extent determined, what types of sensors can be used, and how AUVs can be employed to autonomously and adaptively track these dynamic plumes. Since AUVs vary significantly in design, mobility, deployment duration, on-board processing power, etc., it is also necessary to consider the best choice of AUV (or combination of AUVs) to track a plume. Thus, an operator/scientist's choice of AUV type(s) will likely depend the type of plume to be tracked, or vice versa. Since most underwater plumes are highly spatiotemporally dynamic, employing environmentally adaptive autonomy to track them with a fleet of AUVs is one of the most efficient ways to do so, given today's technology.

An Efficient, Time-of-Flight-Based Underwater Acoustic Ranging System for Small Robotic Fish

Small (decimeter-scale) robotic fish are promising mobile sensor platforms for aquatic environments. Fine-grained localization for dense networks of such robotic fish presents a challenge because of noisy underwater environment, required submeter accuracy, and constraints on onboard processing power and hardware complexity. In this paper, we present an efficient time-of-flight-based acoustic ranging system for localization of robotic fish with limited onboard resources. The system involves simple hardware: a single pair of monotone buzzer and microphone. The distance between two nodes is determined by the time it takes for an acoustic signal generated by the buzzer on the first node to reach the microphone on the second node. The arrival of the signal is detected with the sliding discrete Fourier transform (SDFT) algorithm, where the rise dynamics of the signal is modeled and used for compensation of detection latency. The algorithm is implemented onboard a small biomimetic robotic fish, and experiments in an indoor pool have shown that the compensated SDFT algorithm results in an underwater ranging error of 1.9 wavelengths (1 m), and is thus promising for localization of dense aquatic networks.

Flexible On-Board Stream Processing for Automotive Sensor Data

Vehicle testing and diagnosis requires huge amounts of data to be gathered and analyzed. Not all possibly interesting data can be stored because of the limited memory available in a tested vehicle. On-board preprocessing of data and decisions about which information has to be kept or omitted is thus vital for vehicle testing routines. This paper introduces a method for flexible on-board processing of sensor data of a vehicle. The approach is motivated by sensor network ideas and makes use of stream processing techniques. A processing graph model for automotive applications is proposed, which consists of operator nodes and connecting data streams. This model supplies both recording and processing functionality together. To account for dynamic changes of conditions within a vehicle-most of the time only a small portion of the vehicle states are interesting for diagnosis-both the model and actual software are built in such a way that the whole system can automatically be adapted at runtime whenever certain conditions are detected. The proposed stream processing model has been implemented in a proof-of-concept industrial application, that was deployed to an automotive on-board unit. Results show that this approach effectively trades a little more on-board processing power for a large data volume, that does not need to be saved and transmitted for off-board usage anymore.

Low-Cost Visual Tracking of a Landing Place and Hovering Flight Control with a Microcontroller

The growth of civil and military use has recently promoted the development of unmanned miniature aerial vehicles dedicated to surveillance tasks. These flying vehicles are often capable of carrying only a few dozen grammes of payload. To achieve autonomy for this kind of aircraft novel sensors are required, which need to cope with strictly limited onboard processing power. One of the key aspects in autonomous behaviour is target tracking. Our visual tracking approach differs from other methods by not using expensive cameras but a Wii remote camera, i.e. commodity consumer hardware. The system works without stationary sensors and all processing is done with an onboard microcontroller. The only assumptions are a good roll and pitch attitude estimation, provided by an inertial measurement unit and a stationary pattern of four infrared spots on the target or the landing spot. This paper details experiments for hovering above a landing place, but tracking a slowly moving target is also possible.

Environment and Usage Monitoringof Electronic Products for Health Assessment and Product Design

The life-cycle environmental and usage conditions of a product or system can be monitored and analyzed to assess its on-going health, provide advance warning of failure, and provide information to improve the design and qualification of fielded and future products. The challenge lies in the implementation of this method in application conditions. This paper presents methods to effectively collect and analyze life-cycle environmental and usage data for in-situ health assessments. An integrated hardware-software micro-programmable module for health and usage monitoring of electronic products in their application environment is presented. The hardware incorporates local sensors, and on-board processing power using embedded software. These data processing capabilities are intended to enable immediate and localized processing of the raw sensor data to reduce the power and memory consumption anticipated during load monitoring. Guidelines are also provided to develop a life cycle monitoring plan, that encompasses the selection of environmental and usage parameters. A case study is presented to illustrate the methodology.

Enhanced mission performance from autonomous instrument guidance

During the last decade improvements in electronics, on-board processing power and software design has lead to significant advances in the development of autonomous instrumentation for spacecraft use. The Advanced Stellar Compass (ASC) and the newly developed micro-ASC ( μ ASC ) are excellent examples of such autonomous space instrumentation. With its full autonomy, this star tracker is capable of providing, in real-time, the absolute orientation with respect to the celestial reference frame with an accuracy of a few arc seconds. This high accuracy along with the robust operations, low weight and power consumption makes the μ ASC an ideal instrument for small, high yielding satellite missions. The ASC has hitherto been used by the satellite AOCS and the high accuracy scientific instrument for attitude recovery (among others onboard ØRSTED, CHAMP, and GRACE), and satellite high accuracy target acquisition and pointing (PROBA). Here three applications of the μ ASC as an autonomous onboard precision guide for precision vector instrumentation are presented. These are autonomous onboard antenna guidance, telescope guidance and tracking and high accuracy and wide range laser rangers.

The NRL micro tactical expendable (MITE) air vehicle

AbstractThe US Naval Research Laboratory (NRL) is developing technologies that will enable Navy-relevant missions with the smallest practical Micro Air Vehicles (MAVs). The NRL Micro Tactical Expendable (MITE) air vehicle is a result of this research. MITE is a hand-launched, dual-propeller, fixed-wing air vehicle, with a 25cm chord and a wingspan of 25–47cm, depending on payload weight. Vehicle gross weight is 130–350g. Miniature autopilot systems, based on visual imaging techniques, are being developed for MITE. These will be used in conjunction with conventional autopilot sensors to allow the MITE to fly autonomously. This paper provides an overview of the MITE development, including aerodynamic design considerations, electric propulsion, and vision-based autopilot research. Also presented is a rationale for the development of control laws that can direct the behavior of large groups of MAVs or other vehicle agents. Dubbed ‘physicomimetics,’ this process can bring about the self-assembly of complex MAV formations, though individual MAVs have minimal onboard processing power and limited local sensing capabilities.