Despite significant advances in Large Language Models (LLMs), planning tasks still present challenges for LLM-based agents. Existing planning methods face two key limitations: heavy constraints and cascading errors. To address these limitations, we propose a novel parallel planning paradigm, which Decomposes, Plans for subtasks in Parallel, and Merges subplans into a final plan (DPPM). Specifically, DPPM decomposes the complex task based on constraints into subtasks, generates the subplan for each subtask in parallel, and merges them into a global plan. In addition, our approach incorporates a verification and refinement module, enabling error correction and conflict resolution. Experimental results demonstrate that DPPM significantly outperforms existing methods in travel planning tasks.
This study deals with the problem of task and motion planning of autonomous systems within the context of high-level tasks. Specifically, a task comprises logical requirements (conjunctions, disjunctions, and negations) on the trajectories and final states of agents in certain regions of interest. We propose an optimal planning approach that combines offline computation and online planning. First, a simplified Petri net system is proposed to model the autonomous system. Then, indicating places are designed to implement the logical requirements of the specifications. Building upon this, a compact representation of the state space called extended basis reachability graph is constructed and an efficient online planning algorithm is developed to obtain the optimal plan. It is shown that the most burdensome part of the planning procedure may be removed offline, thanks to the construction of the extended basis reachability graph. Finally, series of simulations are conducted to demonstrate the computational efficiency and scalability of our developed method.
Marišová Eleonóra, Hodossy Kristián, Kováčik Marián
The article explores the interconnection between the reformed Common Agricultural Policy (CAP) for 2023–2027 and the fundamental legislative changes in spatial planning and construction in the Slovak Republic since 2022. The CAP reform introduced an integrated strategic framework that merges direct payments and rural development, requiring alignment of national strategies with the European Union’s environmental and climate objectives. In Slovakia, a parallel reform of spatial planning and construction legislation has taken place, culminating in the adoption of Act No. 200/2022 Coll. on Spatial Planning and the new Construction Act No. 25/2025 Coll. The new legal provisions introduce an integrated procedure for construction intentions, digitalisation of administrative processes, and decentralisation of executive powers through the establishment of new regional offices with extended competences. The authors highlight the benefits of the new system (faster permitting processes, simplified administration) as well as potential risks (tight deadlines for municipalities, legal uncertainty post-2032). They underline the need to harmonise spatial planning tools with CAP objectives, particularly concerning land use, nature conservation, and infrastructure, in order to prevent conflicts between legislation and rural development strategies. The article also aims to analyse the challenges in implementing the environmental goals of the CAP, which are often inadequately fulfilled due to poorly designed eco-schemes and inconsistent funding. The article also includes a comparison of various EU member states’ approaches to CAP implementation.
The aim of the work: to determine the prevalence and establish the frequency of comorbidity in patients with ischemic heart disease when planning myocardial revascularization by performing coronary artery bypass grafting.
Materials and methods. The study included patients with ischemic heart disease who underwent coronary artery bypass grafting on a working heart (n = 3672), both male (n = 3059, 83,3 %) and female (n = 613, 16,7 %), the average age of the study participants was 61.1 ± 0.8 years. The study design consisted of dividing participants into four age groups according to the WHO classification. The material for the analysis was anamnestic and diagnostic and treatment data of 3672 electronic medical histories of patients with ischemic heart disease for the period from 2015 to 2021. All patients included in the study were under dispensary observation at the State Institution “National Institute of Cardiovascular Surgery named after M. M. Amosov NAMS of Ukraine” and at the municipal non-profit enterprise “Cherkasy Regional Cardiology Center of the Cherkasy Regional Council”. To standardize approaches to assessing comorbidity, the Charlson Comorbidity Index (CCI) and the Charlson Age-Augmented Comorbidity Index (CA-CCI) were determined, and a comprehensive assessment of comorbidity was compared with the results of coronary artery bypass grafting.
Results. The frequencies of the most common comorbidities in patients with ischemic heart disease on the eve of myocardial revascularization by coronary artery bypass grafting were determined. In accordance with the study design, taking into account the age of the patients, the frequencies of comorbidities included in the standard deficit of the calculation of the СCI were established. It was determined that the most common comorbidities with ischemic heart disease are: chronic heart failure – 84.0 %, myocardial infarction – 58.7 %, chronic kidney disease – 29.5 % and type 2 diabetes mellitus – 19.7 %. Statistical differences in the frequencies of the most common comorbidities were established taking into account the age of the patients. It was found that the highest CA-CCI was in patients of senile age – 6.8 and in elderly patients – 5.4.
Conclusions. It was determined that the average number of diseases per patient in this sample was 2.6 ± 0.3, and the variation series of the number of existing concomitant diseases ranged from 1 to 7 diseases. The determined average СCI was 3.1 ± 0.3, and the average age-associated СA-СCI was 4.6 ± 0.3, which must be taken into account when planning the method of myocardial revascularization due to the interaction of a separate pathology, which is exacerbated by the aggression of surgical trauma and causes a different spectrum of complications during surgical interventions, increasing perioperative mortality. A comprehensive assessment of the frequency of existing concomitant diseases in a cohort of patients with ischemic heart disease of a large center-based study on the eve of myocardial revascularization by calculating СCI and СA-СCI increases the objectivity of the choice of treatment tactics.
Participatory urban planning is the mainstream of modern urban planning that involves the active engagement of residents. However, the traditional participatory paradigm requires experienced planning experts and is often time-consuming and costly. Fortunately, the emerging Large Language Models (LLMs) have shown considerable ability to simulate human-like agents, which can be used to emulate the participatory process easily. In this work, we introduce an LLM-based multi-agent collaboration framework for participatory urban planning, which can generate land-use plans for urban regions considering the diverse needs of residents. Specifically, we construct LLM agents to simulate a planner and thousands of residents with diverse profiles and backgrounds. We first ask the planner to carry out an initial land-use plan. To deal with the different facilities needs of residents, we initiate a discussion among the residents in each community about the plan, where residents provide feedback based on their profiles. Furthermore, to improve the efficiency of discussion, we adopt a fishbowl discussion mechanism, where part of the residents discuss and the rest of them act as listeners in each round. Finally, we let the planner modify the plan based on residents' feedback. We deploy our method on two real-world regions in Beijing. Experiments show that our method achieves state-of-the-art performance in residents satisfaction and inclusion metrics, and also outperforms human experts in terms of service accessibility and ecology metrics.
Batch planning is increasingly necessary to quickly produce diverse and quality motion plans for downstream learning applications, such as distillation and imitation learning. This paper presents Global Tensor Motion Planning (GTMP) -- a sampling-based motion planning algorithm comprising only tensor operations. We introduce a novel discretization structure represented as a random multipartite graph, enabling efficient vectorized sampling, collision checking, and search. We provide a theoretical investigation showing that GTMP exhibits probabilistic completeness while supporting modern GPU/TPU. Additionally, by incorporating smooth structures into the multipartite graph, GTMP directly plans smooth splines without requiring gradient-based optimization. Experiments on lidar-scanned occupancy maps and the MotionBenchMarker dataset demonstrate GTMP's computation efficiency in batch planning compared to baselines, underscoring GTMP's potential as a robust, scalable planner for diverse applications and large-scale robot learning tasks.
In recent years, the integration of prediction and planning through neural networks has received substantial attention. Despite extensive studies on it, there is a noticeable gap in understanding the operation of such models within a closed-loop planning setting. To bridge this gap, we propose a novel closed-loop planning framework compatible with neural networks engaged in joint prediction and planning. The framework contains two running modes, namely planning and safety monitoring, wherein the neural network performs Motion Prediction and Planning (MPP) and Conditional Motion Prediction (CMP) correspondingly without altering architecture. We evaluate the efficacy of our framework using the nuPlan dataset and its simulator, conducting closed-loop experiments across diverse scenarios. The results demonstrate that the proposed framework ensures the feasibility and local stability of the planning process while maintaining safety with CMP safety monitoring. Compared to other learning-based methods, our approach achieves substantial improvement.
The medium and long-term supply-demand imbalance of the power system in the context of the new power system is becoming more and more prominent due to the fluctuation and intermittency brought about by the high proportion of new energy sources connected to the grid. In this regard, a multi-objective power supply-demand balance optimization method considering the spatiotemporal correlation of source and load is proposed in this work. First, the autocorrelation and inter-correlation characteristics of source and load are analyzed. On this basis, a multi-dimensional scenario set construction method considering the spatiotemporal correlation of source and load is proposed. Then, the planning capacity of each regional power source and the system operation under each scenario are taken as the optimization variables. Renewable energy electricity curtailment, equivalent annual total cost, and inter-region transmission electricity are taken as the optimization objectives. Various constraints such as power source planning and operation, power balance, inter-region power transmission, and renewable energy power curtailment rate are considered comprehensively. The optimization method for the medium and long-term power supply and demand balance is proposed. Finally, the method is applied to Hunan Province, China to guide power planning. The results show that compared with traditional multi-dimensional correlation scene construction methods, the average probability density functions error of wind turbine output, photovoltaic output, and load constructed in this work decrease by 44.08 %, 73.64 %, and 57.54 %, respectively. It takes into account the regional, temporal, temporal autocorrelation, and inter-correlation of the source and load, and has similar characteristics to historical data. Compared with traditional planning that only considers economy, the optimization plan for power supply and demand balance in this work reduces electricity curtailment and inter-region transmission by 97.04 % and 72.71 %, respectively, balancing renewable energy consumption, economy, and regional independent balancing indicators.
With the gradual prominence of global water shortage and other problems, evaluating and predicting the impact of land use change on regional water conservation function is of great reference significance for carrying out national spatial planning and environmental protection, and realizing land intelligent management. We first analyzed 8,416 remote sensing images in the upper reaches of the Hanjiang River Basin (URHRB) by GEE platform and obtained the land use and land cover (LULC) results of fours periods. Through our field investigation, the accuracy of remote sensing image interpretation is obviously higher than that of other comprehensive LULC classification results. Then, through the coupling of InVEST-PLUS model, not only the results of URHRB water conservation from 1990 to 2020 were calculated and the accuracy was assessed, but also the LULC results and water conservation of URHRB under different development scenarios in 2030 were predicted. The results showed as follows: From 1990 to 2020, the forest area of URHRB increased by 7152.23 km2, while the area of cropland, shrub and grassland decreased by 3220.35 km2, 1414.72 km2 and 3385.39 km2, respectively. The InVEST model reliably quantifies the water yield and water conservation of URHRB. In the past 30 years, the total amount of water-saving in China has shown a trend of increasing first and then decreasing. From the perspective of vegetation types, URHRB forest land is the main body of water conservation, with an average annual water conservation depth of 653.87 mm and an average annual water conservation of 472.10×108 m3. Under the ecological protection scenario of the URHRB in 2030, the maximum water conservation in the basin is 574.92×108 m3, but compared with the water conservation in 2010, there is still a gap of 116.28×108 m3. Therefore, through the visualization analysis of the LULC changes of URHRB and water conservation function, it is found that the land and resources department should pay attention to the LULC changes of water sources and adjust the territorial spatial planning in time to cope with the huge water conservation gap in the future.
Temporal abstraction and efficient planning pose significant challenges in offline reinforcement learning, mainly when dealing with domains that involve temporally extended tasks and delayed sparse rewards. Existing methods typically plan in the raw action space and can be inefficient and inflexible. Latent action spaces offer a more flexible paradigm, capturing only possible actions within the behavior policy support and decoupling the temporal structure between planning and modeling. However, current latent-action-based methods are limited to discrete spaces and require expensive planning. This paper presents a unified framework for continuous latent action space representation learning and planning by leveraging latent, score-based diffusion models. We establish the theoretical equivalence between planning in the latent action space and energy-guided sampling with a pretrained diffusion model and incorporate a novel sequence-level exact sampling method. Our proposed method, $\texttt{LatentDiffuser}$, demonstrates competitive performance on low-dimensional locomotion control tasks and surpasses existing methods in higher-dimensional tasks.
Task planning for embodied AI has been one of the most challenging problems where the community does not meet a consensus in terms of formulation. In this paper, we aim to tackle this problem with a unified framework consisting of an end-to-end trainable method and a planning algorithm. Particularly, we propose a task-agnostic method named 'planning as in-painting'. In this method, we use a Denoising Diffusion Model (DDM) for plan generation, conditioned on both language instructions and perceptual inputs under partially observable environments. Partial observation often leads to the model hallucinating the planning. Therefore, our diffusion-based method jointly models both state trajectory and goal estimation to improve the reliability of the generated plan, given the limited available information at each step. To better leverage newly discovered information along the plan execution for a higher success rate, we propose an on-the-fly planning algorithm to collaborate with the diffusion-based planner. The proposed framework achieves promising performances in various embodied AI tasks, including vision-language navigation, object manipulation, and task planning in a photorealistic virtual environment. The code is available at: https://github.com/joeyy5588/planning-as-inpainting.
Excavation plans are crucial in construction projects, dictating the dirt disposal strategy and excavation sequence based on the final geometry and machinery available. While most construction processes rely heavily on coarse sequence planning and local execution planning driven by human expertise and intuition, fully automated planning tools are notably absent from the industry. This paper introduces a fully autonomous excavation planning system. Initially, the site is mapped, followed by user selection of the desired excavation geometry. The system then invokes a global planner to determine the sequence of poses for the excavator, ensuring complete site coverage. For each pose, a local excavation planner decides how to move the soil around the machine, and a digging planner subsequently dictates the sequence of digging trajectories to complete a patch. We showcased our system by autonomously excavating the largest pit documented so far, achieving an average digging cycle time of roughly 30 seconds, comparable to the one of a human operator.
Gender inequality in employment remains a significant challenge in many countries, including South Africa. The impact of this inequality on economic development is a topic of increasing interest and concern, with many studies showing a correlation between gender equality and economic growth. This study explores the impact of gender inequality in employment on economic development in South Africa, with a case study of Nyandeni local municipality. The study employs a quantitative, statistical technique to answer the study issues. The article explores the challenges faced by women in the workforce, the impact of gender employment equity policies, and the role of government, businesses, and civil society in addressing gender inequality in employment and promoting economic development within Nyandeni Local Municipality. This research seeks to determine if women's work will provide an extra lever for economic expansion. A fundamental study on women's involvement in economic development describes the position of women in Africa, Asia, and Latin American nations. The theory that frames the accompanying debate focuses on inequalities, growth, gender, and capacities in general, and the influence of gendered disparities on salaries, education, and economic growth. This research aims to reveal a connection between economic development in South Africa and gender equality. As a second objective, the study seeks to determine whether the contribution of women to economic growth provides extra valuable information for economic policymaking. South Africa's gender gap in employment remains despite legislative and legal progress in the battle against gender inequality.
Hydrogen energy is an important part of China's new energy system,and its market scale and application scenarios are also expanding. Under the guidance of various national policies,provinces and cities have issued relevant policies in the fields of fuel cell vehicles and other fields according to their regional characteristics,which has accelerated the pace of hydrogen energy commercialization to a certain extent. In terms of fuel cells and other aspects,this paper sorted out the national and local hydrogen energy policies,and analyzed the application fields of hydrogen energy,the policy points of the future development planning of provinces and cities,and the profit model. Finally,the corresponding suggestions and prospects were given for the current process of hydrogen energy commercialization,which provide a reference for improving the efficiency of hydrogen energy in the future.
Applications of electric power, Production of electric energy or power. Powerplants. Central stations
The definition and representation of planning problems is at the heart of AI planning research. A key part is the representation of action models. Decades of advances improving declarative action model representations resulted in numerous theoretical advances, and capable, working, domain-independent planners. However, despite the maturity of the field, AI planning technology is still rarely used outside the research community, suggesting that current representations fail to capture real-world requirements, such as utilizing complex mathematical functions and models learned from data. We argue that this is because the modeling process is assumed to have taken place and completed prior to the planning process, i.e., offline modeling for offline planning. There are several challenges inherent to this approach, including: limited expressiveness of declarative modeling languages; early commitment to modeling choices and computation, that preclude using the most appropriate resolution for each action model -- which can only be known during planning; and difficulty in reliably using non-declarative, learned, models. We therefore suggest to change the AI planning process, such that is carries out online modeling in offline planning, i.e., the use of action models that are computed or even generated as part of the planning process, as they are accessed. This generalizes the existing approach (offline modeling). The proposed definition admits novel planning processes, and we suggest one concrete implementation, demonstrating the approach. We sketch initial results that were obtained as part of a first attempt to follow this approach by planning with action cost estimators. We conclude by discussing open challenges.
One of the challenges of task planning is to find out what causes the planning failure and how to handle the failure intelligently. This paper shows how to achieve this. The idea is inspired by the connected graph: each verticle represents a set of compatible \textit{states}, and each edge represents an \textit{action}. For any given initial states and goals, we construct virtual actions to ensure that we always get a plan via task planning. This paper shows how to introduce virtual action to extend action models to make the graph to be connected: i) explicitly defines static predicate (type, permanent properties, etc) or dynamic predicate (state); ii) constructs a full virtual action or a semi-virtual action for each state; iii) finds the cause of the planning failure through a progressive planning approach. The implementation was evaluated in three typical scenarios.
Efficient planning in continuous state and action spaces is fundamentally hard, even when the transition model is deterministic and known. One way to alleviate this challenge is to perform bilevel planning with abstractions, where a high-level search for abstract plans is used to guide planning in the original transition space. Previous work has shown that when state abstractions in the form of symbolic predicates are hand-designed, operators and samplers for bilevel planning can be learned from demonstrations. In this work, we propose an algorithm for learning predicates from demonstrations, eliminating the need for manually specified state abstractions. Our key idea is to learn predicates by optimizing a surrogate objective that is tractable but faithful to our real efficient-planning objective. We use this surrogate objective in a hill-climbing search over predicate sets drawn from a grammar. Experimentally, we show across four robotic planning environments that our learned abstractions are able to quickly solve held-out tasks, outperforming six baselines. Code: https://tinyurl.com/predicators-release
Generalized planning studies the computation of general solutions for a set of planning problems. Computing general solutions with correctness guarantee has long been a key issue in generalized planning. Abstractions are widely used to solve generalized planning problems. Solutions of sound abstractions are those with correctness guarantees for generalized planning problems. Recently, Cui et al. proposed a uniform abstraction framework for generalized planning. They gave the model-theoretic definitions of sound and complete abstractions for generalized planning problems. In this paper, based on Cui et al.'s work, we explore automatic verification of sound abstractions for generalized planning. We firstly present the proof-theoretic characterization for sound abstraction. Then, based on the characterization, we give a sufficient condition for sound abstractions which is first-order verifiable. To implement it, we exploit regression extensions, and develop methods to handle counting and transitive closure. Finally, we implement a sound abstraction verification system and report experimental results on several domains.