Uncluttered Environments Research Articles

Multiphase Autonomous Docking via Model-Based and Hierarchical Reinforcement Learning

With the rise of traffic around Earth’s orbit, spacecraft mission designs have placed an unprecedented demand on the capabilities of autonomous systems. In the early 2000s, the state-of-the-art autonomous spacecraft controllers were designed for static and uncluttered environments. A little over a decade later, the challenges facing spacecraft autonomy now include cluttered, dynamic environments with time-varying constraints, logical modes, fault tolerances, uncertain dynamics, and complex maneuvers. With this rise in complexity, many areas of research have been investigating more experimental control strategies, such as reinforcement learning (RL), as a potential solution to this problem. The research presented herein aims to expand on efforts to quantify the use of RL in autonomous rendezvous, proximity operations, and docking (ARPOD) environments, with consideration to the inherent drawbacks of the more common algorithms present in the field. We present hierarchical model-based RL as a solution to an autonomous docking problem. This algorithm can learn satellite parameters, extrapolate trajectory information, and learn uncertain dynamics via data collection. By using gradient-free model predictive control logic, the algorithm can handle nondifferentiable objectives and complex constraints. Lastly, the hierarchical structure demonstrates an ability to generate feasible trajectories in the presence of integrated third-party subcontrollers commonly found in spacecraft. This study highlights the ability of the hierarchical algorithm to combine and manipulate third-party subpolicies to achieve trajectories not previously trained on.

Multi-View CNN-LSTM Architecture for Radar-Based Human Activity Recognition

In this paper, we propose a Multi-View Convolutional Neural Network and Long Short-Term Memory (CNN-LSTM) network which fuses multiple “views” of the time-range-Doppler radar data-cube for human activity recognition. It adopts the structure of convolutional neural networks to extract optimal frame based features from the time-range, time-Doppler and range-Doppler projections of the radar data-cube. The CNN models are trained using an unsupervised Convolutional Auto-Encoder (CAE) topology. Afterwards, the pre-trained parameters of the encoder are fine-tuned to extract intermediate frame based representations, which are subsequently aggregated via LSTM networks for sequence classification. The temporal correlation among the views is explicitly learned by sharing the LSTM network weights across different views. Moreover, we propose range and Doppler energy dispersion and temporal difference based features as an input to the CNN-LSTM models. Furthermore, we investigate the use of target tracking features as an auxiliary side information. The proposed model is trained on datasets collected in both cluttered and uncluttered environments. For validation, an independent test dataset, with unseen participants, in a cluttered environment was collected. Fusion with auxiliary features improves the generalization by 5%, yielding an overall Macro F1-score of 74.7%.

Wavelength selection approach for an incoherent optical detection sensor (LiDAR)

An energy or direct detection or time-of-flight sensor (a type of incoherent optical detection sensor) used for remote detection and ranging purposes is a useful measurement tool due to its simplicity and high performance in uncluttered environments. A sensor- or top-level design approach has been established [Appl. Opt.59, 1939 (2020)APOPAI0003-693510.1364/AO.384135] due to the usefulness of these sensors, and with this, lower-level designs can be performed to optimize the sensor for particular applications. A critical design element of an incoherent optical detection sensor, or any active optical sensor for that matter, is the selection of a best or optimal central operational wavelength. First and foremost, a relevant metric is developed to provide an optimum wavelength. Then, a search for this wavelength is generated given a generic set of components where conditions are best suited for direct detection sensors, i.e., uncluttered environments or space-like, and finally, the search is again carried out for conditions within the Earth's atmosphere where transmission plays a role.

A Novel Clustering-Based Algorithm for Solving Spatially Constrained Robotic Task Sequencing Problems

The robotic task sequencing problem (RTSP) appears in various forms across many industrial applications and consists of developing an optimal sequence of motions to visit a set of target points defined in a task space. Developing solutions to problems involving complex spatial constraints remains challenging due to the existence of multiple inverse kinematic solutions and the requirements for collision avoidance. So far existing studies have been limited to relaxed RTSPs involving a small number of target points and relatively uncluttered environments. When extending existing methods to problems involving greater spatial constraints and large sets of target points, they either require substantially long planning times or are unable to obtain high-quality solutions. To this end, this article presents a clustering-based algorithm to efficiently address spatially constrained RTSPs involving several hundred to thousands of points. Through a series of benchmarks, we show that the proposed algorithm outperforms the state-of-the-art in terms of solution quality and planning efficiency for large, complex problems, achieving up to 60% reduction in task execution time and 91% reduction in computation time.

Adaptations in the call emission pattern of frugivorous bats when orienting under challenging conditions.

Echolocating bats emit biosonar calls and use echoes arising from call reflections, for orientation. They often pattern their calls into groups which increases the rate of sensory feedback. Insectivorous bats emit call groups at a higher rate when orienting in cluttered compared to uncluttered environments. Frugivorous bats increase the rate of call group emission when they echolocate in noisy environments. In frugivorous bats, it remains unclear if call group emission represents an exclusive adaptation to avoid acoustic interference by signals of conspecifics or if it represents an adaptation that allows to orient under demanding environmental conditions. Here, we compared the emission pattern of the frugivorous bat Carolliaperspicillata when the bats were flying in narrow versus wide or cluttered versus non-cluttered corridors. The bats emitted larger call groups and they increased the call rate within call groups when navigating in narrow or cluttered environments. These adaptations resemble the ones shown when the bats navigate in noisy environments. Thus, call group emission represents an adaptive behavior when the bats orient in complex environments.

Nocturnal navigation by whip spiders: antenniform legs mediate near-distance olfactory localization of a shelter

Navigation by terrestrial arthropods that wander away from a shelter in search of food or mates in uncluttered environments invariably relies on path integration mechanisms, which provide distance and direction information used to compute a straight-line return trajectory. Navigation in cluttered habitats is more typically visually guided, even in species that are nocturnal. Nocturnal, tropical arachnids in the order Amblypygi, however, depend neither on path integration nor on vision to navigate and are hypothesized to instead rely on highly sensitive olfactory structures. The distal articles of their antenniform legs are covered with receptors that presumably bind the odour stimuli thought to control, at least in part, their impressive ability to relocate a shelter after they wander in a rainforest understory. To test the hypothesis that olfactory inputs from the antenniform legs support their near-distance ability to localize a shelter, we conducted a laboratory experiment with the amblypygid Phrynus marginemaculatus in which an olfactory cue (with contact chemoreception precluded) identified the location of a shelter, the position of which was varied as subjects were trained. Localization of a shelter was readily learned and performance was fully retained over a 2-week period in which subjects were not trained. Localization of the odour-cued shelter declined markedly, however, when the distal articles of the antenniform legs were amputated. Moreover, the residual discrimination ability displayed by subjects when they were subsequently retrained depended on how many olfactory receptors had been removed. The results clearly demonstrate that odour cues can support near-distance goal localization and that receptors on the antenniform legs are the locus of the peripheral neural transduction necessary when odour is used to identify the position of a shelter.

Path Planning Generation Algorithm for a Class of UAV Multirotor Based on State of Health of Lithium Polymer Battery

Nowadays, it exists path planning strategies dedicated to generate trajectories considering different navigation issues in UAV multirotors, such as 3D navigation in cluttered and uncluttered environments, obstacle avoidance, and path re-planning. Such path generators are mainly based on the dynamics associated to position and orientation of the UAV, and the attenuation of external disturbances as the wind. However, one of the main limitations of these methods is that they do not take into account the relationship between the path planning task and the energy consumption associated with the battery performance or State of Health (SoH). In this work, a path planning generation algorithm that take into account the evolution of the battery performance is presented. First, the computation of the battery SoH is realized by introducing two degradation models. Subsequently, the path planning algorithm is defined as a multi-objective optimization problem where the objective is to find a feasible trajectory between way-points whiles minimizing the energy consumed and the mission final time depending on the variation of the battery SoH. Finally, the proposed path planning algorithm is compared with a classical path generation method based on polynomial functions to evaluate the minimization of the energy consumption. The simulation results demonstrate that the proposed path planning algorithm is able to generate feasible and minimum energy trajectories despite the constraints in the battery SoH.

Sprint sensitivity and locomotor trade-offs in green anole (Anolis carolinensis) lizards.

How well an organism completes an ecologically relevant task - its performance - is often considered a key factor in determining individual fitness. Historically, ecomorphological studies have examined how morphological traits determine individual performance in a static manner, assuming that differential fitness in a population is due indirectly to differences in morphological traits that determine a simple measure of performance. This assumption, however, ignores many ecological factors that can constrain performance in nature, such as substrate variation and individual behavior. We examined some of these complexities in the morphology-performance-fitness paradigm, primarily the impact that substrate variation has on performance. We measured maximal sprint speed of green anole lizards on four substrates that varied in size and complexity and are used by or available to individuals in nature. Performance decreased significantly from a broad substrate to a narrow substrate, and lizards were three times slower on a complex substrate than the broadest substrate. We also detected trade-offs in running on substrates with different diameters and in cluttered versus uncluttered environments. Furthermore, morphological predictors of performance varied among substrates. This indicates that natural selection may act on different morphological traits, depending on which substrates are used by individuals, as well as an individual's ability to cope with changes in substrate rather than maximal capacities.

Attention to advertising and memory for brands under alcohol intoxication.

In an attempt to discover new possibilities for advertising in uncluttered environments marketers have recently begun using ambient advertising in, for instance, bars and pubs. However, advertising in such licensed premises have to deal with the fact that many consumers are under the influence of alcohol while viewing the ad. This paper examines the effect of alcohol intoxication on attention to and memory for advertisements in two experiments. Study 1 used a forced exposure manipulation and revealed increased attention to logos under alcohol intoxication consistent with the psychopharmacological prediction that alcohol intoxication narrows attention to the more salient features in the visual environment. Study 2 used a voluntary exposure manipulation in which ads were embedded in a magazine. The experiment revealed that alcohol intoxication reduces voluntary attention to ads and leads to a significant reduction in memory for the viewed ads. In popular terms consuming one or two beers reduces brand recall from 40 to 36% while being heavily intoxicated further reduces brand recall to 17%.

Human Movement Behaviour in Urban Spaces: Implications for the Design and Modelling of Effective Pedestrian Environments

Despite a burgeoning research effort directed at the design and modelling of effective urban spaces for pedestrians, remarkably little is known about how pedestrians actually negotiate urban spaces. This paper reports the results of a video-based observational study aimed at exploring: (1) individuals' movement preferences within uncluttered environments, in particular: (a) desired walking speed, (b) microscopic position preferences, and (c) interpersonal distances between companions while walking; and (2) the ways in which these variables might be influenced by the various personal, situational, and environmental factors that characterise the context in which pedestrians move. The microscopic movement trajectories of 2613 participants were investigated in a covert, video-based observational study of three mixed-use (residential/retail) urban environments close to the city centres of Edinburgh and York, United Kingdom. Age, gender, level of mobility, group size, time of day, and location were found to have significant effects on movement preferences across the range of locations studied. We concluded that a number of influential factors affect how humans negotiate urban spaces, and suggested how these factors may be taken into account in attempts to design and model effective urban spaces for pedestrians.

Models and motion planning

We study the complexity of the motion planning problem for a bounded-reach robot in the situation where the n obstacles in its workspace satisfy two of the realistic models proposed in the literature, namely unclutteredness and small simple-cover complexity. We show that the maximum complexity of the free space of a robot with f degrees of freedom in the plane is Θ( n f/2 + n) for uncluttered environments as well as environments with small simple-cover complexity. The maximum complexity of the free space of a robot moving in a three-dimensional uncluttered environment is Θ( n 2 f/3 + n). All these bounds fit nicely between the Θ( n) bound for the maximum free-space complexity for low-density environments and the Θ( n f ) bound for unrestricted environments. Surprisingly—because contrary to the situation in the plane—the maximum free-space complexity is Θ( n f ) for a three-dimensional environment with small simple-cover complexity.

Distributed Algorithms for Multi-Robot Observation of Multiple Moving Targets

An important issue that arises in the automation of many security, surveillance, and reconnaissance tasks is that of observing the movements of targets navigating in a bounded area of interest. A key research issue in these problems is that of sensor placement—determining where sensors should be located to maintain the targets in view. In complex applications involving limited-range sensors, the use of multiple sensors dynamically moving over time is required. In this paper, we investigate the use of a cooperative team of autonomous sensor-based robots for the observation of multiple moving targets. In other research, analytical techniques have been developed for solving this problem in complex geometrical environments. However, these previous approaches are very computationally expensive—at least exponential in the number of robots—and cannot be implemented on robots operating in real-time. Thus, this paper reports on our studies of a simpler problem involving uncluttered environments—those with either no obstacles or with randomly distributed simple convex obstacles. We focus primarily on developing the on-line distributed control strategies that allow the robot team to attempt to minimize the total time in which targets escape observation by some robot team member in the area of interest. This paper first formalizes the problem (which we term CMOMMT for i>Cooperative Multi-Robot Observation of Multiple Moving Targets) and discusses related work. We then present a distributed heuristic approach (which we call A-CMOMMT) for solving the CMOMMT problem that uses weighted local force vector control. We analyze the effectiveness of the resulting weighted force vector approach by comparing it to three other approaches. We present the results of our experiments in both simulation and on physical robots that demonstrate the superiority of the A-CMOMMT approach for situations in which the ratio of targets to robots is greater than 1/2. Finally, we conclude by proposing that the CMOMMT problem makes an excellent domain for studying multi-robot learning in inherently cooperative tasks. This approach is the first of its kind for solving the on-line cooperative observation problem and implementing it on a physical robot team.

An opportunistic global path planner

In this paper we describe a robot path-planning algorithm that constructs a global skeleton of free-space by incremental local methods. The curves of the skeleton are the loci of maxima of an artificial potential field that is directly proportional to distance of the robot from obstacles. Our method has the advantage of fast convergence of local methods in uncluttered environments, but it also has a deterministic and efficient method of escaping local extremal points of the potential function. We first describe a general roadmap algorithm, for configuration spaces of any dimension, and then describe specific applications of the algorithm for robots with two and three degrees of freedom.