Hasil untuk "Regional planning"

Mir Md Sajid Sarwar

The recent advancement in artificial intelligence (AI) technologies facilitates a paradigm shift toward automation. Autonomous systems are fully or partially replacing manually crafted ones. At the core of these systems is automated planning. With the advent of powerful planners, automated planning is now applied to many complex and safety-critical domains, including smart energy grids, self-driving cars, warehouse automation, urban and air traffic control, search and rescue operations, surveillance, robotics, and healthcare. There is a growing need to generate explanations of AI-based systems, which is one of the major challenges the planning community faces today. The thesis presents a comprehensive study on explainable artificial intelligence planning (XAIP) for hybrid systems that capture a representation of real-world problems closely.

Alessandro Burigana, Francesco Fabiano

Epistemic planning extends (multi-agent) automated planning by making agents' knowledge and beliefs first-class aspects of the planning formalism. One of the most well-known frameworks for epistemic planning is Dynamic Epistemic Logic (DEL), which offers an rich and natural semantics for modelling problems in this setting. The high expressive power provided by DEL make DEL-based epistemic planning a challenging problem to tackle both theoretically, and in practical implementations. As a result, existing epistemic planners often target different DEL fragments, and typically rely on ad hoc languages to represent benchmarks, and sometimes no language at all. This fragmentation hampers comparison, reuse, and systematic benchmark development. We address these issues by introducing the Epistemic Planning Domain Definition Language (EPDDL). EPDDL provides a unique PDDL-like representation that captures the entire DEL semantics, enabling uniform specification of epistemic planning tasks. Our main contributions are: 1. A formal development of abstract event models, a novel representation for epistemic actions used to define the semantics of our language; 2. A formal specification of EPDDL's syntax and semantics grounded in DEL with abstract event models. Through examples of representative benchmarks, we illustrate how EPDDL facilitates interoperability, reproducible evaluation, and future advances in epistemic planning.

C. D. S. Gamboa, Ma. R. C. O. Ang, J. M. Medina

Wind energy is the most abundant, mature, and wide range of all renewable energy, and its optimal use results in enhanced energy security and affordability. However, wind farm suitability analysis is a crucial decision-making process that involves various stakeholders with different biases and concerns. Thus, effective planning of limited land resources and utilization of renewables ensure low environmental impact, high economic gains, and minimal community disturbances. This study presents a GIS-based AHP-PROMETHEE II framework that addresses the need for efficient onshore wind farms suitability analysis. The framework is illustrated through a case study in Cagayan Province. The results showed that there are 10.16% unsuitable, 47.45% least suitable, and 6.08% most suitable areas in the study area. The results also showed that the least suitable areas are inland while the most suitable areas are near coastal areas as validated by the sole wind energy service contract in the study area. In addition, the AHP’s underlying value judgments, inconsistencies, and limitations can be addressed by PROMETHEE II. The proposed framework can be used as a more strategic planning tool by decision-makers for onshore wind farm siting developments and can be a guide for regional or national scale mapping.

Afagh Mehri Shervedani, Matthew R. Walter, Milos Zefran

Recent advances in large language models (LLMs) have demonstrated their potential as planners in human-robot collaboration (HRC) scenarios, offering a promising alternative to traditional planning methods. LLMs, which can generate structured plans by reasoning over natural language inputs, have the ability to generalize across diverse tasks and adapt to human instructions. This paper investigates the potential of LLMs to facilitate planning in the context of human-robot collaborative tasks, with a focus on their ability to reason from high-level, vague human inputs, and fine-tune plans based on real-time feedback. We propose a novel hybrid framework that combines LLMs with human feedback to create dynamic, context-aware task plans. Our work also highlights how a single, concise prompt can be used for a wide range of tasks and environments, overcoming the limitations of long, detailed structured prompts typically used in prior studies. By integrating user preferences into the planning loop, we ensure that the generated plans are not only effective but aligned with human intentions.

Emanuele La Malfa, Ping Zhu, Samuele Marro et al.

We present an end-to-end framework for planning supported by verifiers. An orchestrator receives a human specification written in natural language and converts it into a PDDL (Planning Domain Definition Language) model, where the domain and problem are iteratively refined by sub-modules (agents) to address common planning requirements, such as time constraints and optimality, as well as ambiguities and contradictions that may exist in the human specification. The validated domain and problem are then passed to an external planning engine to generate a plan. The orchestrator and agents are powered by Large Language Models (LLMs) and require no human intervention at any stage of the process. Finally, a module translates the final plan back into natural language to improve human readability while maintaining the correctness of each step. We demonstrate the flexibility and effectiveness of our framework across various domains and tasks, including the Google NaturalPlan benchmark and PlanBench, as well as planning problems like Blocksworld and the Tower of Hanoi (where LLMs are known to struggle even with small instances). Our framework can be integrated with any PDDL planning engine and validator (such as Fast Downward, LPG, POPF, VAL, and uVAL, which we have tested) and represents a significant step toward end-to-end planning aided by LLMs.

Adnan Jafar, Xun Jia

Current radiation therapy treatment planning is limited by suboptimal plan quality, inefficiency, and high costs. This perspective paper explores the complexity of treatment planning and introduces Human-Centric Intelligent Treatment Planning (HCITP), an AI-driven framework under human oversight, which integrates clinical guidelines, automates plan generation, and enables direct interactions with operators. We expect that HCITP will enhance efficiency, potentially reducing planning time to minutes, and will deliver personalized, high-quality plans. Challenges and potential solutions are discussed.

Hamid Amoonia, Mohammadreza Yousefi Roshan, Amir Hasan Jangi

Understanding vegetation spatial patterns is vital for assessing ecosystem health, especially in water-stressed arid and semi-arid regions. These areas, often characterized by sparse and sensitive vegetation, face significant challenges from climate change and human pressures. Rashtkhar County in Iran exemplifies such an environment, experiencing a severe decline in groundwater levels, which necessitates precise monitoring of its vegetation dynamics. Quantifying the spatial dispersion or heterogeneity of vegetation cover objectively is challenging. While indices like NDVI provide information on vegetation density, they often miss crucial details about spatial structure and fragmentation. Shannon's entropy model, derived from information theory, offers a robust method to measure the complexity or dispersion within a spatial system. Specifically, Shannon's relative entropy (G), a normalized index, allows for standardized comparisons of dispersion across time and space, effectively differentiating between uniform and highly fragmented landscapes. Given the ecological importance of vegetation in Rashtkhar and the intense pressure on its water resources, this study's primary objective was to utilize Shannon's relative entropy model to quantitatively assess the trend of spatial vegetation cover dispersion from 1990 to 2023. A key goal was to analyze the relationship between this dispersion trend, groundwater level fluctuations (approx. 1986-2022), and the region's topography. The findings aim to provide insights for sustainable resource management in similar vulnerable areas, highlighting the need for quantitative tools to understand complex ecosystem dynamics.This research employed a descriptive-analytical approach using remote sensing and GIS to investigate quantitative trends in vegetation spatial dispersion in Rashtkhar County, a predominantly arid area in Khorasan Razavi Province, Iran (~4,360 km²). The study period covered 1990-2023 for vegetation and water years 1365-66 to 1400-1401 (approx. 1986-2022) for groundwater. Time-series Landsat imagery (TM and OLI sensors, 30m resolution) for 1990, 2000, 2010, 2015, 2020, and 2023 were acquired from USGS. After standard radiometric and atmospheric corrections, NDVI was calculated using ENVI 5.6 to map vegetation cover for each year. Groundwater level data were sourced from the provincial Regional Water Authority. To incorporate topography, the area was classified into five elevation zones using a DEM (Figure 3). Shannon's relative entropy model was then applied to quantify spatial dispersion annually. Relative entropy (G) was calculated as G = H / ln(n), where H = - Σ [pi * ln(pi)], 'pi' is the proportion of vegetation cover in topographic zone 'i', and 'n' is the number of zones (n=5). G ranges from 0 (maximum concentration) to 1 (maximum dispersion). Calculations were performed using Excel and ArcGIS. Finally, the temporal trend of G was compared with groundwater level trends to analyze the interplay between vegetation spatial structure and water resource availability.The analysis revealed significant and contrasting trends in groundwater levels and vegetation dispersion in Rashtkhar County. Groundwater elevation showed a severe and nearly continuous decline over the ~35-year period, dropping approximately 50 meters from above 1110m to just over 1060m. This drastic reduction highlights critical pressure on groundwater resources and aquifer depletion. Conversely, Shannon's relative entropy (G) for vegetation cover, measuring spatial dispersion, followed a different trajectory. G increased sharply from 0.801 in 1990 to 0.913 in 2000. Despite minor fluctuations, it maintained a high level with a slight upward trend, reaching 0.924 in 2023. This indicates a sustained shift towards greater heterogeneity and spatial dispersion of vegetation cover. While the total vegetation area mapped via NDVI showed an overall increase (from ~8,000 ha to nearly 30,000 ha), its spatial distribution changed, with increased contributions from low and high elevation zones at the expense of mid-elevation areas. The key finding is the stark contrast: severe groundwater depletion occurred concurrently with increased (or stabilization at high levels of) vegetation spatial dispersion. This seemingly paradoxical outcome likely does not signify ecological improvement. Instead, it strongly suggests a complex structural rearrangement of the landscape driven by adaptation to water stress. The increased dispersion might result from factors like shifts towards more drought-resistant but potentially scattered crops (e.g., saffron), fragmentation of agricultural lands due to changing economic viability or land tenure, uneven development, or the patchy survival of native vegetation in micro-refugia. The analysis also confirmed that topography significantly influences how this dispersion manifests, modulating the vegetation system's response to environmental pressures like water scarcity.This study quantitatively assessed vegetation spatial dispersion in Rashtkhar County using Shannon's relative entropy, relating it to groundwater trends and topography. A major finding was the contrasting long-term trends: a severe, continuous ~50-meter decline in groundwater levels (approx. 1986-2022) alongside a significant increase and stabilization of vegetation spatial dispersion (G rising from 0.801 in 1990 to 0.924 in 2023). This increased heterogeneity, occurring amidst resource depletion, is interpreted not as ecological recovery but as evidence of landscape structural rearrangement. This rearrangement likely reflects adaptive responses to water stress, such as changes in land use, cropping patterns towards more scattered cultivation, land fragmentation, or patchy vegetation survival. Topography was confirmed as a key factor modulating these spatial patterns. The research demonstrates the utility of Shannon's relative entropy as a valuable quantitative tool for capturing complex spatial dynamics and their interactions with environmental drivers like water availability in arid/semi-arid regions. It highlights that relying solely on overall vegetation indices can be misleading, and understanding spatial structure is crucial for assessing ecosystem sustainability. These findings underscore the need to incorporate such spatial metrics into sustainable resource management frameworks, particularly for developing integrated water conservation and land use planning strategies in fragile ecosystems under pressure. Future work could enhance this understanding by incorporating detailed land use data, grazing information, and higher-resolution remote sensing inputs.

Atharva Gundawar, Mudit Verma, Lin Guan et al.

As the applicability of Large Language Models (LLMs) extends beyond traditional text processing tasks, there is a burgeoning interest in their potential to excel in planning and reasoning assignments, realms traditionally reserved for System 2 cognitive competencies. Despite their perceived versatility, the research community is still unraveling effective strategies to harness these models in such complex domains. The recent discourse introduced by the paper on LLM Modulo marks a significant stride, proposing a conceptual framework that enhances the integration of LLMs into diverse planning and reasoning activities. This workshop paper delves into the practical application of this framework within the domain of travel planning, presenting a specific instance of its implementation. We are using the Travel Planning benchmark by the OSU NLP group, a benchmark for evaluating the performance of LLMs in producing valid itineraries based on user queries presented in natural language. While popular methods of enhancing the reasoning abilities of LLMs such as Chain of Thought, ReAct, and Reflexion achieve a meager 0%, 0.6%, and 0% with GPT3.5-Turbo respectively, our operationalization of the LLM-Modulo framework for TravelPlanning domain provides a remarkable improvement, enhancing baseline performances by 4.6x for GPT4-Turbo and even more for older models like GPT3.5-Turbo from 0% to 5%. Furthermore, we highlight the other useful roles of LLMs in the planning pipeline, as suggested in LLM-Modulo, which can be reliably operationalized such as extraction of useful critics and reformulator for critics.

Brandon Vu, Toki Migimatsu, Jeannette Bohg

Task and Motion Planning (TAMP) algorithms solve long-horizon robotics tasks by integrating task planning with motion planning; the task planner proposes a sequence of actions towards a goal state and the motion planner verifies whether this action sequence is geometrically feasible for the robot. However, state-of-the-art TAMP algorithms do not scale well with the difficulty of the task and require an impractical amount of time to solve relatively small problems. We propose Constraints and Streams for Task and Motion Planning (COAST), a probabilistically-complete, sampling-based TAMP algorithm that combines stream-based motion planning with an efficient, constrained task planning strategy. We validate COAST on three challenging TAMP domains and demonstrate that our method outperforms baselines in terms of cumulative task planning time by an order of magnitude. You can find more supplementary materials on our project \href{https://branvu.github.io/coast.github.io}{website}.

Faruq Al Muqsit, Rukuh Setiadi, Alex Lo

Waste management is a complex challenge for cities in developing countries, including Indonesia. This issue is particularly severe in heritage tourism areas, where unmanaged waste can affect the aesthetics and image of historic places. Therefore, this study aimed to investigate the sustainability aspects of waste management using heritage visitors’ destinations and waste management operator areas of the Old City of Semarang. A quantitative method was used to achieve this objective, and data were collected through questionnaires, interviews, field observations, document reviews and analyzed using descriptive methods. The results showed that visitors had positive behavior and awareness regarding waste management. Furthermore, sustainable waste management was implemented but not fully realized across all aspects. This study offered various measures to improve sustainable waste management in area, including waste sorting, collaboration between stakeholders, and policy advocacy on sustainable waste management.

T. Beatley

Yuxiao Chen, Sushant Veer, Peter Karkus et al.

In highly interactive driving scenarios, the actions of one agent greatly influences those of its neighbors. Planning safe motions for autonomous vehicles in such interactive environments, therefore, requires reasoning about the impact of the ego's intended motion plan on nearby agents' behavior. Deep-learning-based models have recently achieved great success in trajectory prediction and many models in the literature allow for ego-conditioned prediction. However, leveraging ego-conditioned prediction remains challenging in downstream planning due to the complex nature of neural networks, limiting the planner structure to simple ones, e.g., sampling-based planner. Despite their ability to generate fine-grained high-quality motion plans, it is difficult for gradient-based planning algorithms, such as model predictive control (MPC), to leverage ego-conditioned prediction due to their iterative nature and need for gradient. We present Interactive Joint Planning (IJP) that bridges MPC with learned prediction models in a computationally scalable manner to provide us the best of both the worlds. In particular, IJP jointly optimizes over the behavior of the ego and the surrounding agents and leverages deep-learned prediction models as prediction priors that the join trajectory optimization tries to stay close to. Furthermore, by leveraging homotopy classes, our joint optimizer searches over diverse motion plans to avoid getting stuck at local minima. Closed-loop simulation result shows that IJP significantly outperforms the baselines that are either without joint optimization or running sampling-based planning.

Heba Mohamed Ammar, Kariman Ahmed Shawkyorcid

تتسم المدن بالتغير المستمر بشكل تكيفي لتتجاوز الأزمات الطبيعية والمجتمعية التي تواجهها، يتم هذا التغير والتطور باستحداث آليات متنوعة لمواكبة التغير المستمر. وشهدت مصر سلسلة من التغيرات والتحديات الاقتصادية الاجتماعية، ومن أحدث ما واجهته مصر والعالم جائحة فيروس كورونا المستجد (كوفيد-19). وتعد المجتمعات ذات الدخل المنخفض الأكثر تأثراً بالأزمات والتغيرات والأقل قدره على معالجة التغيرات والتكيف معها، ووفقاً لتقديرات البنك الدولي لعام 2020 بلغت نسبة السكان ذوي الدخل المنخفض بمصر نحو 26.1%. تعتمد زيادة قدرة المجتمعات وخاصة ذات الدخل المنخفض على التكيف ومواجهة الأزمات على أسباب متعددة، أحد أهمها مدى فعالية منظومة ادارة العمران. وتساهم الحوكمة التكيفية في تمكين المجتمعات المحلية للتصدي للأزمات سواء المحلية أو العالمية، وفي تحقيق المرونة وتعزيز قدرة المجتمع على التكيف بفاعلية مع المتغيرات المختلفة. لذا يهدف البحث للتوصل لكيفية دمج أبعاد الحوكمة التكيفية بمنظومة إدارة العمران بمؤسسات التنمية العمرانية بمصر خاصة لمشروعات تنمية المجتمعات ذات الدخل المنخفض. وتحديد أوجه القوة التي تساهم في إمكانية تطبيقها وأوجه القصور الواجب مراعاتها وتحسينها. للوصول لذلك الهدف يقوم البحث بالتعرف على مفهوم الحوكمة التكيفية وأبعادها والنماذج النظرية المتعلقة بها بالمراجعة التحليلية للدراسات النظرية، بالإضافة إلى تحليل نماذج من التجارب العالمية المطبقة لمعايير الحوكمة التكيفية، لتحديد مدى امكانية تطبيق تلك المعايير بمؤسسات التنمية العمرانية بمصر، كما تم اجراء استبيان للخبراء بمجال التخطيط العمراني وإدارة العمران لتحديد أولويات تطبيق معايير الحوكمة التكيفية بالسياق المصري. وتتلخص نتائج البحث في الوصول لإطار لتطبيق الحوكمة التكيفية بمؤسسات التنمية العمرانية بمصر، والذي قد يساهم في تحسين أوضاع المجتمعات ذات الدخل المنخفض لتكون اكثر مرونة وقدرة على التكيف والصمود ضمن الإطار المؤسسي لإدارة العمران بمصر. Cities are constantly adapting to deal with the natural and societal crises they are experiencing. This adaptation requires developing various mechanisms to keep pace with ongoing changes. Egypt has experienced a range of social and economic changes and challenges. (COVID-19) is the most recent crisis in Egypt and the whole world. Low-income communities are the most affected by crises and changes while being the least able to adapt to them, furthermore, Egypt's low-income population was about 26.1% according to the World Bank estimation for 2020. Achieving more resilient communities, particularly low-income, is dependent on multiple factors. The efficiency of the urban management system is one of the most crucial factors, Adaptive governance empowers communities to respond to both local and global crises, increases resilience, and enhances society's ability to adapt to changing factors. The research aims to examine how to integrate the adaptive governance aspects in the management system of Egyptian urban development institutions, especially for low-income community development projects. In addition, identify the strengths and weaknesses of applying adaptive governance. Based on the analytical theoretical approach the research reviews literature of adaptive governance definition, aspects, and theoretical models. International case studies have been selected and analyzed to determine the applicability of adaptive governance indicators to Egyptian urban development institutions. In addition, an expert questionnaire was conducted as an empirical study, to examine ranking of the adaptive governance indicators according to priority in the Egyptian context. The paper ends with a framework for adaptive governance application in the Egyptian urban development institutions. Which could be applied and enhanced the resilience of the low-income communities.

Tom Barry, Tom Barry, Tom Christensen et al.

The Arctic Council working group, the Conservation of Arctic Flora and Fauna (CAFF) established the Circumpolar Biodiversity Monitoring Programme (CBMP), an international network of scientists, governments, Indigenous organizations, and conservation groups working to harmonize and integrate efforts to extend and develop monitoring and assessment of the Arctic’s biodiversity. Its relevance stretches beyond the Arctic to a broad range of regional and global initiatives and agreements. This paper describes the process and approach taken in the last two decades to develop and implement the CBMP. It documents challenges encountered, lessons learnt, and solutions, and considers how it has been a model for national, regional, and global monitoring programmes; explores how it has impacted Arctic biodiversity monitoring, assessment, and policy and concludes with observations on key issues and next steps. The following are overarching prerequisites identified in the implementation of the CBMP: effective coordination, sufficient and sustained funding, improved standards and protocols, co-production of knowledge and equitable involvement of IK approaches, data management to facilitating regional analysis and comparisons, communication and outreach to raising awareness and engagement in the programme, ensuring resources to engage in international fora to ensuring programme implementation.

Shan Gao, Chen Chang, Qiang Liu et al.

The evacuation in the university building complex involving a variety of campus buildings and a high density of occupants is barely simulated, especially by considering the safety awareness of evacuees and the impact of measured stair evacuation speed. This paper simulates and optimizes the fire emergency evacuation scheme of a university building complex in Northwest China by using BIM technology and Pathfinder simulation evacuation software. The proposed evacuation optimization scheme of university building complex reduces the evacuation time by 15–20% compared with the original evacuation scheme of the university. Through the simulation, it is found that the evacuation time required for sleeping at night is the shortest, while the longest evacuation time is required during lunch, due to the dispersion of students. Therefore, in addition to considering the safety awareness of the evacuees and the measured evacuation speed of stairs, the corresponding campus evacuation plan should be formulated according to the distribution of buildings in the campus complex and the actual utilization in different periods, so as to avoid the safety exit congestion and being far away from the refuge, due to the dense population in a specific period of time. This study also provides a method for solving the problem of regional evacuation path planning.

Yongtao Kan, Li Zhang, Yan Wang et al.

Strain Q11^T of an irregular coccoid Gram-positive bacterium, aerobic and non-motile, was isolated from <i>Meconopsis integrifolia</i> seeds. Strain Q11^T grew optimally in 1% (<i>w</i>/<i>v</i>) NaCl, pH 7, at 30 °C. Strain Q11^T is most closely related to <i>Flexivirga</i>, as evidenced by 16S rRNA gene analysis, and shares the highest similarity with <i>Flexivirga aerilata</i> ID2601S^T (99.24%). Based on genome sequence analysis, the average nucleotide identity and digital DNA–DNA hybridization values of strains Q11^T and D2601S^T were 88.82% and 36.20%, respectively. Additionally, strain Q11^T showed the abilities of nitrogen fixation and indole acetic acid production and was shown to promote maize growth under laboratory conditions. Its genome contains antibiotic resistance genes (the vanY gene in the vanB cluster and the vanW gene in the vanI cluster) and extreme environment tolerance genes (ectoine biosynthetic gene cluster). Shotgun proteomics also detected antibiotic resistance proteins (class A beta-lactamases, D-alanine ligase family proteins) and proteins that improve plant cold tolerance (multispecies cold shock proteins). Strain Q11^T was determined to be a novel species of the genus <i>Flexivirga</i>, for which the name <i>Flexivirga meconopsidis</i> sp. nov. is proposed. The strain type is Q11^T (GDMCC 1.3002^T = JCM 36020 ^T).

Wenjia Zhang, Chenyu Fang, Lin Zhou et al.

Abstract Understanding the spatial structure of a megaregion, like the Pearl River Delta (PRD) or the emerging Guangdong-Hong Kong-Macao Greater Bay Area in China, is important for regional planning and governance. However, few studies have conceptualized varying regional spatial structure via the network perspective. This study thus develops a complex network approach, particularly by adopting a novel machine-learning-based weighted stochastic block model and visual analytics, to measuring potential spatial mesoscale structures in PRD. We build a finer-grained commuting network with 60 sub-city divisions as nodes by aggregating a large set of mobile phone signaling data collected in 2018. Results detect a hybrid polycentric configuration of two community components, each with a core-periphery structure inside. One community centered on Guangzhou has a semi-core commuting belt alongside the intercity railway and high-speed rails, while the other centered on Shenzhen exhibits a concentric commuting ring. Intercity transport infrastructure, spatial proximity, and regional integration policies appear to play important roles in shaping spatial and network structures in the megaregion, while the constraint of administrative boundary is dissolving. This study finally discusses the implication for regional coordination policies and spatial planning strategies in PRD and the Greater Bay Area.

Raúl Trujillo-Cabezas, J. Verdegay

The regional information program specializes in spatial development, economics, land and housing economics, sustainable planning, information management and marketing in the agriculture/food business. research and education in spatial economics, regional and development finance, spatial development and planning, urban and rural information, spatial analysis, Management Information Systems (MIS) of agricultural and food industries, e-Business, Information management and marketing in the food business. Students in the regional information program can study understand, analyze, and propose theories for urban and regional structures in terms of the regional and spatial economy and information systems. Students also learn to use diverse statistical and econometric tools such as Geographic Information Systems (GIS) and Management Information Systems (MIS) for analyzing regional/agricultural information and food business.

Vivek Veeriah, Zeyu Zheng, Richard Lewis et al.

Planning with a learned model is arguably a key component of intelligence. There are several challenges in realizing such a component in large-scale reinforcement learning (RL) problems. One such challenge is dealing effectively with continuous action spaces when using tree-search planning (e.g., it is not feasible to consider every action even at just the root node of the tree). In this paper we present a method for selecting affordances useful for planning -- for learning which small number of actions/options from a continuous space of actions/options to consider in the tree-expansion process during planning. We consider affordances that are goal-and-state-conditional mappings to actions/options as well as unconditional affordances that simply select actions/options available in all states. Our selection method is gradient based: we compute gradients through the planning procedure to update the parameters of the function that represents affordances. Our empirical work shows that it is feasible to learn to select both primitive-action and option affordances, and that simultaneously learning to select affordances and planning with a learned value-equivalent model can outperform model-free RL.

Felipe Orellana, Joaquín Hormazábal, Gonzalo Montalva et al.

Coastal areas concentrate a large portion of the country’s population around urban areas, which in subduction zones commonly are affected by drastic tectonic processes, such as the damage earthquakes have registered in recent decades. The seismic cycle of large earthquakes primarily controls changes in the coastal surface level in these zones. Therefore, quantifying temporal and spatial variations in land level after recent earthquakes is essential to understand shoreline variations better, and to assess their impacts on coastal urban areas. Here, we measure the coastal subsidence in central Chile using a multi-temporal differential interferometric synthetic aperture radar (MT-InSAR). This geographic zone corresponds to the northern limit of the 2010 Maule earthquake (Mw 8.8) rupture, an area affected by an aftershock of magnitude Mw 6.8 in 2019. The study is based on the exploitation of big data from SAR images of Sentinel-1 for comparison with data from continuous GNSS stations. We analyzed a coastline of ~300 km by SAR interferometry that provided high-resolution ground motion rates from between 2018 and 2021. Our results showed a wide range of subsidence rates at different scales, of analyses on a regional scale, and identified the area of subsidence on an urban scale. We identified an anomalous zone of subsidence of ~50 km, with a displacement <−20 mm/year. We discuss these results in the context of the impact of recent earthquakes and analyze the consequences of coastal subsidence. Our results allow us to identify stability in urban areas and quantify the vertical movement of the coast along the entire seismic cycle, in addition to the vertical movement of coast lands. Our results have implications for the planning of coastal infrastructure along subduction coasts in Chile.