Stewart Worrall

A probabilistic method for detecting impending vehicle interactions

In mining operations it is advantageous to be able to predict the future movements of nearby vehicles. For autonomous mining, this can be used for localised, short term path planning and risk assessment. For semi-autonomous or non-autonomous mining, this can be used for collision avoidance, situational awareness and risk assessment of maneuvers between a human operated vehicle, and another vehicle (operated by a human or otherwise). This paper introduces a probabilistic approach to predicting vehicle movements, in particular, the time until two vehicle paths intersect. Results are shown using real data collected from the operation of two separate fleets of vehicles.

Using Non-Parametric Filters and Sparse Observations to Localise a Fleet of Mining Vehicles

Mining operations generally involve a large number of expensive vehicles, and for the efficient management of these vehicles it is very beneficial to know their location at all times. The current procedure for vehicle localisation in mines is to provide the mine with complete wireless network coverage to facilitate the broadcasting of vehicle positions. This paper examines an alternative method of localisation that does not require the expense of a radio network with full mine coverage. Two different non-parametric filter approaches are presented to estimate the location of the vehicles. A comparison of the two filters is also presented with experimental results using data collected in two operational mines.

Haul truck alignment monitoring and operator warning system

This paper presents a haul truck alignment monitoring system that provides early warning signals to the operator when the truck is about to lose control. It is considered that a truck is in this condition when it crosses the center of the road or veers to the side of the road at speed in an uncontrolled manner. The system provides different levels of warnings to the driver and other haul trucks in visual contact. The system is based on a laser range and bearing sensor that measure the relative distance to standard polyvinyl chloride poles located at the side of the road. Using this approach, a high level of accuracy and reliability can be achieved with low cost since the installation and maintenance of the infrastructure does not require special expertise or expensive machinery. The system also logs raw sensor data and warning events generated by the truck. This information is downloaded using wireless interfaces to a base station for postprocessing purposes. This is essential to monitor the operation of the system and determine potential degradation of its components. Experimental results are shown demonstrating the robustness of the system. These results were obtained from actual data extracted from a database built with more that 12 months of continued operation of the system in two different mines.

Vehicle collision probability calculation for general traffic scenarios under uncertainty

Vehicle-to-vehicle (V2V) communication systems allow vehicles to share state information with one another to improve safety and efficiency of transportation networks. One of the key applications of such a system is in the prediction and avoidance of collisions between vehicles. If a method to do this is to succeed it must be robust to measurement uncertainty. The method should also be general enough that it does not rely on constraints on vehicle motion for the accuracy of its predictions. It should work for all interactions between vehicles and not just a select subset. This paper presents a method for collision probability calculation that addresses these problems.

A Context-Based Approach to Vehicle Behavior Prediction

Despite the best efforts of research and development carried out in the automotive industry, accidents continue to occur resulting in many deaths and injuries each year. It has been shown that the vast majority of accidents occur as a result (at least in part) of human error. This paper introduces the model for the Intelligent Systems for Risk Assessment (ISRA) project which has the goal of eliminating accidents by detecting risk, alerting the operators when appropriate, and ultimately removing some control of the vehicle from the operator when the risk is deemed unacceptable. The underlying premise is that vehicle dynamic information without contextual information is insufficient to understand the situation well enough to enable the analysis of risk. This paper defines the contextual information required to analyze the situation and shows how location context information can be derived using collected vehicle data. The process to infer high level vehicle state information using context information is also presented. The experimental results demonstrate the context based inference process using data collected from a fleet of mining vehicles during normal operation. The systems developed for the mining industry can later be extended to include more complex traffic scenarios that exist in the domain of ITS.

Improving vehicle safety using context based detection of risk

When mining vehicle operators take risks, there is a increased probability of an accident that can cause injuries, fatalities and large financial costs to the mine operators. It can be assumed that the operators do not intentially take unnecessarily high risk, and that the risks are hidden due to factors such as adverse weather, fatigue, visual obstructions, boredom, etc. This paper examines the potential of measuring the risk of danger in a situation by using the safe rules of operation defined by mining safety management. The problem with measuring safety is that the safe rules of operation are heavily dependent on the context of the situation. What is considered normal practice and safe in one part of the mine (such as performing a u-turn in a parking lot) is not safe on a haul road. To be able to measure safety, it is therefore necessary to understand the different context areas in a mine so that feedback of unsafe behaviour presented to the operator is relevant to the context of the situation. This paper presents a novel method for generating context area information using the vehicle trajectory information collected from vehicles in the mine. Results are presented using real-life data collected from several operating fleets of mining vehicles.

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.

The Warrigal Dataset: Multi-Vehicle Trajectories and V2V Communications

Intelligent Transportation Systems rely on understanding, predicting and affecting the interactions between vehicles. Development of such systems must be based upon data derived from actual interactions if they are to be effective when used in real world applications. Increasingly, systems are being developed that are based on radio communication of state and intent between vehicles. Understanding of how these interactions occur is also necessary to creating robust systems. In order to test and compare new techniques, approaches and algorithms it is necessary to have a rich dataset to experiment with. This paper presents a detailed dataset useful for members of the Intelligent Transportation Systems community. It contains vehicle state information, vehicle-to-vehicle communications and road maps at high temporal resolution for large numbers of interacting vehicles over a long time period. This data set has already been used for a number of Intelligent Transportation Systems projects such as road mapping, driver intent prediction and collision avoidance among others.

Using Delayed Observations for Long-Term Vehicle Tracking in Large Environments

The tracking of vehicles over large areas with limited position observations is of significant importance in many industrial applications. This paper presents algorithms for long-term vehicle motion estimation based on a vehicle motion model that incorporates the properties of the working environment and information collected by other mobile agents and fixed infrastructure collection points. The prediction algorithm provides long-term estimates of vehicle positions using speed and timing profiles built for a particular environment and considering the probability of a vehicle stopping. A limited number of data collection points distributed around the field are used to update the estimates, with negative information (no communication) also used to improve the prediction. This paper introduces the concept of observation harvesting, a process in which peer-to-peer communication between vehicles allows egocentric position updates to be relayed among vehicles and finally conveyed to the collection point for an improved position estimate. Positive and negative communication information is incorporated into the fusion stage, and a particle filter is used to incorporate the delayed observations harvested from vehicles in the field to improve the position estimates. The contributions of this work enable the optimization of fleet scheduling using discrete observations. Experimental results from a typical large-scale mining operation are presented to validate the algorithms.

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.