Matt Garratt

UNSW Sydney | UNSW

Algorithms using swarming collective motion can solve coverage problems in unknown environments by reacting to unknown obstacles in real-time when they are encountered. However, these algorithms face two key challenges when deployed on real robots. First, hand-tuning efficient collective motion parameters is both time-consuming and difficult. Secon...

This is the post-print version of the paper accepted for 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). Note: © 20XX IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for adverti...

This paper introduces a state‐machine model designed for a multi‐modal, multi‐robot environmental sensing algorithm tailored to dynamic real‐world settings. The multi‐modal algorithm uniquely combines two distinct exploration strategies for gas source localization and mapping tasks: (1) an initial exploration phase using multi‐robot coverage path p...

At present the perception system of autonomous vehicles is grounded on 3D vision technologies along with deep learning to process depth information. Although deep learning models for 3D perception give promising results, recent research demonstrates that they are also vulnerable to adversarial attacks just like deep learning models trained on 2D im...

This paper introduces a novel approach for improving gas source localization in dynamic urban environments, employing a swarm of nano-Crazyflie drones through a hybrid strategy that integrates Adaptive Robotic Particle Swarm Optimization (ARPSO) with Bidirectional Brain Emotional Learning (BBEL). The proposed method refines the ARPSO algorithm into...

This paper introduces a state-machine model for a multi-modal, multi-robot environmental sensing algorithm tailored to dynamic real-world settings. The algorithm uniquely combines two exploration strategies for gas source localization and mapping: (1) an initial exploration phase using multi-robot coverage path planning with variable formations for...

Nano helicopter unmanned air vehicles (NHUAVs) offer significant advantages due to their compact size, maneuverability , and cost-effectiveness, making them versatile and flexible platforms for various applications. Despite these merits, achieving full autonomy in NHUAV control presents challenges including multi-variable, strongly coupled nonlinea...

Insects are excellent at flying in dense vegetation and navigating through other complex spatial environments. This study investigates the strategies used by honeybees (Apis mellifera) to avoid collisions with an obstacle encountered frontally during flight. Bees were trained to fly through a tunnel that contained a solitary vertically oriented cyl...

Autonomous blimps have potential applications in surveillance, monitoring, and advertising. Due to their lower payload capacities and possible unstable flight profile, small blimps have been mostly used in indoor applications. However, recent advancements in their design and control have increased the prospects of deploying them for outdoor applica...

Ensuring robust and precise tracking control in the presence of uncertain multi-input–multi-output (MIMO) system dynamics and environmental variations is a significant challenge in the field of robust and adaptive control theory. While fuzzy control strategies have demonstrated good tracking performance in normal conditions, designing and tuning fu...

Miniature blimps are lighter-than-air vehicles which have become an increasingly common unmanned aerial system research platform due to their extended endurance and collision tolerant design. The UNSW-C bio-inspired miniature blimp consists of a 0.5m spherical mylar envelope filled with helium. Four fins placed along the equator provide control ove...

Deep learning networks have demonstrated outstanding performance in 2D and 3D vision tasks. However, recent research demonstrated that these networks result in failures when imperceptible perturbations are added to the input known as adversarial attacks. This phenomenon has recently received increased interest in the field of autonomous vehicles an...

This paper presents an evolutionary framework for generating diverse libraries of collective motion behaviours. It builds upon recent advancements in machine recognition of collective motion and the transformation of random motions into structured collective behaviours. The paper describes the design of the framework, including the use of a fitness...

The novel framework for estimating dense scene flow using depth camera data is demonstrated in the article. Using these estimated flow vectors to identify obstacles improves the path planning module of the autonomous vehicle's (AV) intelligence. The primary difficulty in the development of AVs has been thought to be path planning in cluttered envir...

This paper proposes an innovative approach to coverage path planning and obstacle avoidance for multiple Unmanned Ground Vehicles (UGVs) in a changing environment, taking into account constraints on the time, path length, number of UGVs and obstacles. Our approach leverages deformable virtual leader-follower formations to enable UGVs to adapt their...

Beginning with navigation system design, this paper presents a comprehensive strategy for enhancing the search and rescue capabilities of agile mobile robots. Towards this, the autonomous ground vehicle (AGV) utilizes surface classification to determine and prioritize the terrain it is traversing. Our developed system design incorporates real-time...

The behaviour of social insects such as bees and ants has influenced the development of swarm robots. To enable robots to cooperate together, swarm robotics employs principles such as communication, coordination, and collaboration. Collaboration among multiple robots can lead to a faster task completion time compared to the utilisation of a single,...

Adversarial Machine Learning (AML) represents the ability to disrupt Machine Learning (ML) algorithms through a range of methods that broadly exploit the architecture of deep learning optimisation. This paper presents Distributed Adversarial Regions (DAR), a novel method that implements distributed instantiations of computer vision-based AML attack...

A self-organising bidirectional fuzzy brain emotional learning (SO-BFBEL) controller is developed to control a quadcopter UAV in an uncertain environment. The proposed SO-BFBEL controller improves the performance of the existing BFBEL controller by generating more accurate fuzzy layers in real-time and removes the need to depend on expert knowledge...

This paper proposes a state-machine model for a multi-modal, multi-robot environmental sensing algorithm. This multi-modal algorithm integrates two different exploration algorithms: (1) coverage path planning using variable formations and (2) collaborative active sensing using multi-robot swarms. The state machine provides the logic for when to swi...

Robot swarms are becoming popular in domains that require spatial coordination. Effective human control over swarm members is pivotal for ensuring swarm behaviours align with the dynamic needs of the system. Several techniques have been proposed for scalable human–swarm interaction. However, these techniques were mostly developed in simple simulati...

This paper proposes an iterative transfer learning approach to achieve swarming collective motion in groups of mobile robots. By applying transfer learning, a deep learner capable of recognizing swarming collective motion can use its knowledge to tune stable collective motion behaviors across multiple robot platforms. The transfer learner requires...

This paper presents the development of a type-2 evolving fuzzy control system (T2-EFCS) to facilitate self-learning (either from scratch or from a certain predefined rule). Our system has two major learning stages, namely, structure learning and parameters learning. The structure phase does not require previous information about the fuzzy structure...

A common assumption of coverage path planning research is a static environment. Such environments require only a single visit to each area to achieve coverage. However, some real-world environments are characterised by the presence of unexpected, dynamic obstacles. They require areas to be revisited periodically to maintain an accurate coverage map...

Recent work has shown the possibility for groups of robots to self-bootstrap collective motion behaviours in open arenas. However, for real-world swarm robotics missions, such as package delivery, the environment is likely to be cluttered with obstacles and the swarm will have a target goal. This paper proposes an architecture for self-bootstrappin...

Depth estimation from a single RGB image has attracted great interest in autonomous driving and robotics. State-of-the-art methods are usually designed on top of complex and extremely deep network architectures, which require more computational resources. Moreover, the inherent characteristic of the backbone used by the existing approaches results...

Multi-Gas source localisation and mapping is a challenging problem because multiple measurements must be taken to ensure accurate localisation. This paper presents a novel flocking control strategy for multi-robot exploration and gas field mapping to address this problem. The algorithm includes an active sensing mechanism for driving a flock of age...

As an essential component for many autonomous driving and robotic activities such as ego-motion estimation, obstacle avoidance and scene understanding, monocular depth estimation (MDE) has attracted great attention from the computer vision and robotics communities. Over the past decades, a large number of methods have been developed. To the best of...

The use of the ‘ship as a wave buoy analogy’ (SAWB) provides a novel means to estimate sea states, where relationships are established between causal wave properties and vessel motion response information. This study focuses on a model-free machine learning approach to SAWB-based sea state estimation (SSE), using neural networks (NNs) to map vessel...

Monocular depth estimation is an important task that can be applied to many robotic applications. Existing methods focus on improving depth estimation accuracy via training increasingly deeper and wider networks, however these suffer from large computational complexity. Recent studies found that edge information are important cues for convolutional...

This paper proposes a novel swarm-based control algorithm for exploration and coverage of unknown environments, while maintaining a formation that permits short-range communication. The algorithm combines two elements: swarm rules for maintaining a close-knit formation and frontier search for driving exploration and coverage. Inspired by natural sy...

Quadrotors are one of the popular unmanned aerial vehicles (UAVs) due to their versatility and simple design. However, the tuning of gains for quadrotor flight controllers can be laborious, and accurately stable control of trajectories can be difficult to maintain under exogenous disturbances and uncertain system parameters. This article introduces...

While the concept of swarm intelligence was introduced in 1980s, the first swarm optimisation algorithm was introduced a decade later, in 1992. In this paper, nineteen representative original swarm optimisation algorithms are analysed to extract their common features and design a taxonomy for swarm optimisation. We use twenty-nine benchmark problem...

The use of the `ship as a wave buoy analogy' (SAWB) provides a novel means to estimate sea states, where relationships are established between causal wave properties and vessel motion response information. This study focuses on a model-free machine learning approach to SAWB-based sea state estimation (SSE), using neural networks (NNs) to map vessel...

Collective behaviours such as swarm formation of autonomous agents offer the advantages of efficient movement, redundancy, and potential for human guidance of a single swarm organism. However, tuning the behaviour of a group of agents so that they swarm, is difficult. Behaviour-bootstrapping algorithms permit agents to self-tune behaviour adapted f...

Recent work has developed value functions that can recognize emergent swarming behaviour and distinguish it from random behaviour. To date, this work has been done in point-mass swarm simulations. This paper proposes a transfer learning approach that can improve the performance of a value system for recognising swarming in simulated and real robots...

Quadrotors are one of the popular unmanned aerial vehicles (UAVs) due to their versatility and simple design. However, the tuning of gains for quadrotor flight controllers can be laborious, and accurately stable control of trajectories can be difficult to maintain under exogenous disturbances and uncertain system parameters. This paper introduces a...

Image based Localization (IbL) uses both Structure from Motion (SfM) and Simultaneous Localization and Mapping (SLAM) data for accurate pose estimation. However, under conditions where there is a large perspective difference between the SfM images and SLAM keyframes, the SfM-SLAM co-visibility graph becomes sparse. As a result, the scale drift can...

This paper presents an algorithm for a decentralized self-organizing map. With the explosion in the availability of robotics platforms, and their increasing application to multi-agent systems and robot swarms, there is a need for a new generation of machine learning algorithms that can exploit the distributed nature of sensing and processing that c...

Path planning for multiple Unmanned Ground Vehicles (UGVs) is a critical problem for UGV autonomy and is increasingly attracting attention due to its wide applications. This paper presents a continuous ant colony-based multi-UGV path planner, which consists of UGV path planning and multi-UGV coordination. A continuous Ant Colony Optimisation with a...

View Video Presentation: https://doi.org/10.2514/6.2022-1834.vid Miniature autonomous blimps are autonomous lighter-than-air vehicles that offer a variety of benefits over other existing flight platforms. In particular, blimps offer long flight times, soft envelopes that are resilient to collisions, and friendly human-robot interaction opportunitie...

In recent years, increases in industrial residue have become a significant environmental threat. These residues can cause problems for natural ecosystems and their inhabitants, including animals and humans. Environmental monitoring through sensing is one approach to predict or detect the presence of pollution from such residue. One of the emerging...

In this paper, a new model-free Model-Actor (MA) reinforcement learning controller is developed for output feedback control of a class of discrete-time systems with input saturation constraints. The proposed controller is composed of two neural networks, namely a model-network and an actor network. The model-network is utilized to predict the outpu...

In mobile robotics, area exploration and coverage are critical capabilities. In most of the available research, a common assumption is global, long-range communication and centralised cooperation. This paper proposes a novel swarm-based coverage control algorithm that relaxes these assumptions. The algorithm combines two elements: swarm rules and f...

Depth is a vital piece of information for autonomous vehicles to perceive obstacles. Due to the relatively low price and small size of monocular cameras, depth estimation from a single RGB image has attracted great interest in the research community. In recent years, the application of Deep Neural Networks (DNNs) has significantly boosted the accur...

As an essential component for many autonomous driving and robotic activities such as ego-motion estimation, obstacle avoidance and scene understanding, monocular depth estimation (MDE) has attracted great attention from the computer vision and robotics communities. Over the past decades, a large number of methods have been developed. To the best of...

A Multi-operator continuous Ant Colony Optimisation (MACOR) is proposed in this paper to solve the real-world problems. An adaptive multi-operator framework is proposed for selecting the suitable operator during different evolutionary stages by considering the historical performance of operators and the convergence status of the population. To impr...

This paper aims to design an enhanced self-adaptive interval type-2 fuzzy control system (ESAF2C) for stabilization of a quadcopter drone under external disturbances. Due to the ability to accommodate the footprint-of-uncertainty (FoU), an interval type-2 Takagi-Sugeno fuzzy scheme is employed to directly address the uncertainties in the nonlinear...

Several artificial intelligence-based systems have been proposed to help human operators in safety-critical domains such as air-traffic control systems, health and farming. In scenarios where multiple autonomous systems are utilised, the cognitive capacity of humans hits its limits quickly, and sometimes the success of these systems is obstructed....

Leveraging the benefits of the multiobjective particle swarm optimization (PSO) technique, we introduce a new concept of adaptive strictly negative imaginary (SNI) controllers. The proposed adaptive control systems are specifically designed to minimize a certain performance index, representing the objective of our control design, which is to obtain...

In this paper, an effective approach for real-time multi-obstacle detection and tracking in the navigation module is discussed.Francis, SobersAnavatti, Sreenatha G.Garratt, MatthewAbbass, Hussein A. To calculate a feasible path for an autonomous ground vehicle (AGV) from the start position to goal position, efficient Dstar lite global planner is ad...

Quadrotor system is subject to multiple disturbances, including both internal and external effects (e.g. wind gusts, coupling effects, and unmodeled dynamics). For example, severe wind disturbances may significantly degrade trajectory tracking during the flight of autonomous aerial vehicles, or even cause loss of control or failure of a tracking mi...

This paper presents a compelling combination of the two negative-imaginary (NI) control systems: a consensus-based formation control framework for a hybrid multi-vehicle system and a new dynamic obstacle detection and avoidance algorithm. The incorporation of the two techniques permits robots to self-detect collisions and then self-create a safe pa...

Uninhabited aerial vehicles (UAVs) are widely used in many areas for completing complex missions such as tracking targets, search and rescue, farming (shepherding in the traditional sense) and mapping. Mission planning for shepherding a UAV swarm is advantageous for Human-Swarm Teaming. While most research on shepherding see the shepherd as a simpl...

A novel bidirectional fuzzy brain emotional learning (BFBEL) controller is proposed to control a class of uncertain nonlinear systems such as the quadcopter unmanned aerial vehicle (QUAV). The proposed BFBEL controller is non-model based and has a simplified fuzzy neural network structure and a novel bidirectional brain emotional learning (BBEL) al...

This chapter presents the applications of an interval Type-2 (IT2) Takagi-Sugeno (T-S) fuzzy system for modeling and control the dynamics of a quad-copter unmanned aerial vehicle (UAV). In addition to being complex and non-linear, the dynamics of a quadcopter are under-actuated and uncertain, making the modeling and control tasks across its full fl...

Robotic aircraft are often required to operate in harsh environments (e.g., underground mining, cluttered environments, and battlefields). In this chapter, we discuss an adaptive (evolving) fuzzy system that has the ability to learn and to configure itself based on the human way of learning, which is also somewhat akin to the principles of natural...

This chapter presents an effective approach to path planning combined with task assignments for a group of unmanned aerial vehicles (UAVs). Path planning for a UAV is a challenging task. Handling multiple UAVs in dynamic environments makes planning more complicated. On the other hand, coordination and cooperation is very significant for multi-UAV p...

This chapter aims to demonstrate how rule-based Artificial Intelligence algorithms can address a few human swarm teaming challenges. We will start from the challenges identified by the cognitive engineering community for building human autonomy teaming and how they scale to human swarm teaming. The discussion will follow with a description of rule-...

Shepherding is a specific class of flocking behaviour where external agents (the shepherd) influence the movements of a group of agents (the flock). In nature, a powerful example is herding a flock of sheep by an influential sheepdog. When the sheepdog encroaches on the sheep’s influence zone, the sheep is essentially “repelled”. Optimising this ph...

A novel technique has been developed for autonomous swarm-based unknown environment scouting. A control method known as swarm shepherding was employed, which replicates the behaviour seen when a sheepdog guides a herd of sheep to an objective location. The guidance of the swarm agents was implemented using low computation cost, force-based behaviou...

Developmental evolution of collective swarm behaviours promises new ways to evolve swarms with different movement characteristics. Preliminary work has developed value functions that can recognize emergent swarm behaviour and distinguish it from random behaviour in point-mass boid simulations. This paper examines the performance of several variants...

Wind gusts are a significant barrier to the outdoor operation of teams of Unmanned Aerial Vehicle (UAVs) which operate in close proximity to each other or obstacles. In these situations, traditional control methods such as PID control may not perform adequately. Based on the Strictly Negative Imaginary (SNI) systems theory, this paper presents a no...

Recently, Type-2 fuzzy systems have become increasingly prominent as they have been applied to various nonlinear control applications. This article presents an adaptive fuzzy controller based on the sliding-mode control theory. The proposed self-adaptive interval Type-2 fuzzy controller (SAF2C) is based on the Takagi-Sugeno (TS) fuzzy model and it...