Asher Bender

Autonomous exploration of large-scale benthic environments

Maturing technology has allowed the reliable deployment of robots into large-scale environments for monitoring and exploration applications. Planning techniques which ignore the value of information gathered during transit are able to operate efficiently in these environments and generate trajectories between specified starting and ending locations. Including the value of information gathered during transit increases the complexity of the problem and often leads to algorithms which are unable to scale up to large environments. This paper presents a method for planning informative surveys in large-scale unexplored environments. The proposed methodology does not require a starting or ending location as a constraint. Instead, robot operators are required to specify a survey template, which satisfies both vehicle constraints and the scientific objectives of the deployment. This constraint converts the exploration problem into an experimental design problem where the objective is to choose a location for the specified survey trajectory. A functional representation of the survey utility is learnt using a Gaussian process. This model allows the utility of candidate survey placements to be queried in a continuous space and in arbitrary locations. The proposed exploration method is demonstrated and validated on marine data. The objective is to design a survey which allows the spatial distribution of habitats in a large marine environment to be estimated accurately. The results show that the proposed exploration method is able to model the hidden survey utility function successfully and recommend informative survey placements.

Benthic monitoring with robotic platforms — The experience of Australia

Australias Integrated Marine Observing System (IMOS) has a strategic focus on the impact of major boundary currents on continental shelf environments, ecosystems and biodiversity. To improve our understanding of natural, climate change, and human-induced variability in shelf environments, the IMOS Autonomous Underwater Vehicle (AUV) facility has been charged with generating physical and biological observations of benthic variables that cannot be cost-effectively obtained by other means. Starting in 2010, the IMOS AUV facility began collecting precisely navigated benthic imagery using AUVs at selected reference sites on Australias shelf. This observing program capitalizes on the unique capabilities of AUVs that have allowed repeated visits to the reference sites, providing a critical observational link between oceanographic and benthic processes. This paper provides a brief overview of the relevant capabilities of the AUV facility, the design of the IMOS benthic sampling program, and some preliminary results. We also report on some of the challenges and potential benefits to be realized from a benthic observation system that collects several TB of geo-referenced stereo imagery a year. This includes collaborative semi-automated image analysis, clustering and classification, large scale visualization and data mining, and lighting correction for change detection and characterization. We also mention some of the lessons from operating an AUV-based monitoring program and future work in this area.

An Unsupervised Approach for Inferring Driver Behavior From Naturalistic Driving Data

Intelligent transportation systems are able to collect large volumes of high-resolution data. The amount of data collected by these systems can quickly overwhelm the ability of human analysts to draw meaningful conclusions from the data, particularly in large-scale multivehicle field trials. As advanced driver assistance systems develop, they will also be required to form a rich and high-level understanding of the world from the data they receive, including the behavior of the driver. These applications motivate the need for unsupervised tools capable of forming a high-level summary of low-level driving data. This paper presents an unsupervised method for converting naturalistic driving data into high-level behaviors. The proposed method works in two steps. In the first step, inertial data are automatically decomposed into linear segments. In the second step, the segments are assigned to high-level driving behaviors. The proposed method is computationally efficient and completely unsupervised and requires minimal preprocessing. Although the method is unsupervised, the clusters produced exhibit high-level patterns that can easily be associated with driving behaviors such as braking, turning, accelerating, and coasting. The effectiveness of the proposed algorithms is demonstrated in an offline application where the objective is to summarize inertial data into driving behaviors. The method is also demonstrated in an online application where the aim is to infer the current driving behavior using only inertial data. Both experiments were conducted using driving data collected in natural driving conditions.

Predicting Driver Intent from Models of Naturalistic Driving

Modern advanced driver assistance systems (ADAS) have lead to safer vehicles. However, current ADAS are typically limited to a reactive, physical model of the vehicle. They lack the ability to understand complex traffic scenarios. One traffic scenario that has gathered interest in recent years is the problem of inferring driver behaviour at road features such as intersections. At these locations drivers may choose to perform one of many available manoeuvres. Early identification of the manoeuvre is important for the development of future safety and situational awareness systems. The objective of this paper is to develop a method for predicting which manoeuvre a driver will execute. To fulfil this objective a simple method based on quadratic discriminant analysis is proposed. The method is computationally efficient and developed with a view to being applied to complex road networks using naturalistic driving data. The proposed method is demonstrated and validated using naturalistic driving data collected at a three way T-intersection.

Classification with probabilistic targets

Modern robotic platforms, deployed for environmental monitoring and mapping, are able to rapidly accumulate large data sets. Whilst the data sets collected by these platforms are highly descriptive, they are often too large for human experts to analyse exhaustively. Although the large data sets could be analysed by humans in principle, the amount of labour and time required to process them is not cost effective. In this paper we focus on the classification task of learning the relationship between low resolution, remotely sensed data and categories derived from direct observations of the same phenomenon. To reduce the labour requirements of categorising the direct observations we forgo human supervision and rely on an unsupervised clustering model to segregate the observations into similar groups of data. Rather than using the discrete cluster labels to train a conventional classifier, we develop a new Gaussian process classifier capable of accepting probabilistic training targets. This allows the probabilistic information generated during clustering to be preserved during classification. We demonstrate the new model, in an environmental monitoring application, using data collected by an autonomous underwater vehicle.

Adaptive exploration of benthic habitats using Gaussian processes

Currently, the majority of AUV missions follow fixed pre-programmed surveys. In exploration missions, the environment is unknown and pre-programmed surveys risk wasting limited resources on data with little scientific value. This risk can be mitigated by allowing autonomous agents to adapt their behaviour to suit the environment and the scientific goals of the survey. This paper presents a method for performing adaptive surveys which combines elements from the fields of perception, machine learning and planning. During exploration, a Gaussian mixture model is used to classify sensor data. The classes returned by the Gaussian mixture model are modelled spatially using a Gaussian process classifier. This spatial model is used to guide the agents exploration into informative areas of the environment using value iteration. The advantage of using adaptive surveys and its potential for outperforming pre-programmed surveys is demonstrated in an example application.

Analysis of Propulsion Methods for Long-Range AUVs

Underwater gliders use a buoyancy engine and symmetric wings to produce lift. During operation, gliders follow a saw-tooth trajectory, making them useful vehicles for profiling ocean chemistry. By operating at low speeds with low hotel loads, gliders achieve a high endurance. Man-portable, propeller-driven autonomous underwater vehicles (AUVs) are capable of level flight and can also follow terrain to yield high-quality benthic imagery. These platforms typically operate at high speeds with high hotel loads resulting in relatively low endurance. Although both vehicles are used to collect oceanographic data, constraints on how these vehicles are used differentiate the nature of data they collect. This article examines whether one method of propulsion can provide an intrinsic advantage in terms of horizontal range at low speed, regardless of sampling design. The authors employ first-principle analysis that concludes that either class of vehicle can be designed to achieve the same horizontal transit performance regardless of speed. This result implies that the choice of propulsion method should be driven exclusively by the application and operational requirements.

Multimodal information-theoretic measures for autonomous exploration

Autonomous underwater vehicles (AUVs) are widely used to perform information gathering missions in unseen environments. Given the sheer size of the ocean environment, and the time and energy constraints of an AUV, it is important to consider the potential utility of candidate missions when performing survey planning. In this paper, we utilise a multimodal learning approach to capture the relationship between in-situ visual observations, and shipborne bathymetry (ocean depth) data that are freely available a priori. We then derive information-theoretic measures under this model that predict the amount of visual information gain at an unobserved location based on the bathymetric features. Unlike previous approaches, these measures consider the value of additional visual features, rather than just the habitat labels obtained. Experimental results with a toy dataset and real marine data demonstrate that the approach can be used to predict the true utility of unexplored areas.

GPS/GNSS Consistency in a Multi-path Environment and During Signal Outages

The majority of Intelligent Transportation System (ITS) applications require an estimate of position, often generated through the fusion of satellite based positioning (such as GPS) with on-board inertial systems. To make the position estimates consistent it is necessary to understand the noise distribution of the information used in the estimation algorithm. For GNSS position information the noise distribution is commonly approximated as zero mean with Gaussian distribution, with the standard deviation used as an algorithm tuning parameter. A major issue with satellite based positioning is the well known problem of multipath which can introduce a non-linear and non-Gaussian error distribution for the position estimate. This paper introduces a novel algorithm that compares the noise distribution of the GNSS information with the more consistent noise distribution of the local egocentric sensors to effectively reject GNSS data that is inconsistent. The results presented in this paper show how the gating of the GNSS information in a strong multipath environment can maintain consistency in the position filter and dramatically improve the position estimate. This is particularly important when sharing information from different vehicles as in the case of cooperative perception due to the requirement to align information from various sources.

A Flexible System Architecture for Acquisition and Storage of Naturalistic Driving Data

Innovation in intelligent transportation systems relies on analysis of high-quality data. In this paper, we describe the design principles behind our data management infrastructure. The principles we adopt place an emphasis on flexibility and maintainability. This is achieved by breaking up code into a modular design that can be run on many independent processes. Message passing over a publish-subscribe network enables interprocess communication and promotes data-driven execution. By following these principles, rapid prototyping and experimentation with new sensing modalities and algorithms are possible. The communication library underpinning our proposed architecture is compared against several popular communication libraries. Features designed into the system make it decentralized, robust to failure, and amenable to scaling across multiple machines with minimal configuration. Code written using the proposed architecture is compact, transparent, and easy to maintain. Experimentation shows that our proposed architecture offers a high performance when compared against alternative communication libraries.