This paper presents a path planning methodology which enables Autonomous Underwater Vehicles (AUVs) to navigate in shallow complex environments such as coral reefs. The approach leverages prior information from an aerial photographic survey, and derived bathymetric information of the corresponding area. From these prior maps, a set of features is obtained which define an expected arrangement of objects and bathymetry likely to be perceived by the AUV when underwater. A navigation graph is then constructed by predicting the arrangement of features visible from a set of test points within the prior, which allows the calculation of the shortest paths from any pair of start and destination points. A maximum likelihood function is defined which allows the AUV to match its observations to the navigation graph as it undertakes its mission. To improve robustness, the history of observed features are retained to facilitate possible recovery from non-detectable or misclassified objects. The approach is evaluated using a photo-realistic simulated environment, and results illustrate the merits of the approach even when only a relatively small number of features can be identified from the prior map.

This paper presents two novel approaches for improving image-based underwater obstacle detection by combining sparse stereo point clouds with monocular semantic image segmentation. Generating accurate image-based obstacle maps in cluttered underwater environments, such as coral reefs, are essential for robust robotic path planning and navigation. However, these maps can be challenged by factors including visibility, lighting and dynamic objects (e.g. fish) that may lead to falsely identified free space or dynamic objects which trajectory planners may react to undesirably. We propose combining feature-based stereo matching with learning-based segmentation to produce a more robust obstacle map. This approach considers direct binary learning of the presence or absence of underwater obstacles, as well as a multiclass learning approach to classify their distance (near, mid and far) in the scene. An enhancement to the binary map is also shown by including depth information from sparse stereo matching to produce 3D obstacle maps of the scene. The performance is evaluated using field data collected in cluttered, and at times, visually degraded coral reef environments. The results show improved image-wide obstacle detection, rejection of transient objects (such as fish), and range estimation compared to feature-based sparse and dense stereo point clouds alone.

Genetic algorithm-based route choice optimization techniques have recently demonstrated their potential to reduce the travel costs on the system level. In this context, the costs are reduced according to exactly one metric such as, e. g., the sum of all individual travel times. In the present study, we will investigate the tradeoff between the saved travel time and the additional costs in terms of higher emissions and a higher fuel

This paper focuses on a parallel Genetic Algorithm approach to solve the coordinated route planning problem in road navigation. The optimization problem is solved locally in each car, local solutions are exchanged when communication opportunities arise. Such Island Genetic Algorithms with local sub-populations are known to yield good results in many domains. Here, we leverage their power for road traffic, where limited knowledge and limited communication abilities make a distributed solution particularly desirable. To this end, we specifically tackle the question how to represent the optimization problem and solution instances, and how to exchange them between cars.

Traffic assignment is a complex optimization problem. In case the road network has many links (thus a high number of alternative routes) and multiple origin-destination pairs, most existing solutions approximate the so-called user equilibrium (a variant of Nash equilibrium). Furthermore, the quality of these solutions (mostly, iterative algorithms) come at the expense of computational performance. In this study, we introduce a methodology to evaluate an approximation of an optimal traffic assignment from the global network's perspective based on genetic algorithms. This approach has been investigated in terms of both network performance (travel time) and convergence speed.

Traffic assignment is an important stage in traffic modeling. Most of the existing approaches are based on finding an approximate solution to the user equilibrium or to the system optimum, which can be computationally expensive. In this paper we use a genetic algorithm to compute an approximate solution (routes for the trips) that seeks to minimize the average travel time. To illustrate this approach, a non-trivial network is used, departing from binary route choice scenarios. Our result shows that the proposed approach is able to find low travel times, without the need of recomputing shortest paths iteratively.

A car navigation system’s job is to plan a good route from an origin to a destination. There are many different options how this can be accomplished. Path choices can be calculated based on static road map representations, or they can take into account dynamic information like, e. g., known road blocks or the current traffic situation. More recently, the idea has gained ground that navigation systems could even cooperate in order to co-ordinate route choices so as to proactively avoid the formation of traffic jams. While several heuristics for algorithms to improve the vehicles’ route choices have been proposed, little is known about the potential benefit of such optimizations. How much can we gain if dynamic information exchange and/or co-ordination between vehicles are used? Answering this question requires to obtain information on the travel times realized by “best possible”, globally co-ordinated route choices—and therefore the solution of a highly complex optimization problem. Here, we propose a method to accomplish this. We use genetic algorithm optimization to jointly evolve the route choices of all cars in a street network iteratively towards an optimal solution, where the quality of each intermediate optimization step is assessed using a road traffic simulation.