Path Selection Research Articles

Bi-directional adaptive enhanced A* algorithm for mobile robot navigation

PurposeThe present paper aims to address challenges associated with path planning and obstacle avoidance in mobile robotics. It introduces a pioneering solution called the Bi-directional Adaptive Enhanced A* (BAEA*) algorithm, which uses a new bidirectional search strategy. This approach facilitates simultaneous exploration from both the starting and target nodes and improves the efficiency and effectiveness of the algorithm in navigation environments. By using the heuristic knowledge A*, the algorithm avoids unproductive blind exploration, helps to obtain more efficient data for identifying optimal solutions. The simulation results demonstrate the superior performance of the BAEA* algorithm in achieving rapid convergence towards an optimal action strategy compared to existing methods.Design/methodology/approachThe paper adopts a careful design focusing on the development and evaluation of the BAEA* for mobile robot path planning, based on the reference [18]. The algorithm has remarkable adaptability to dynamically changing environments and ensures robust navigation in the context of environmental changes. Its scale further enhances its applicability in large and complex environments, which means it has flexibility for various practical applications. The rigorous evaluation of our proposed BAEA* algorithm with the Bidirectional adaptive A* (BAA*) algorithm [18] in five different environments demonstrates the superiority of the BAEA* algorithm. The BAEA* algorithm consistently outperforms BAA*, demonstrating its ability to plan shorter and more stable paths and achieve higher success rates in all environments.FindingsThe paper adopts a careful design focusing on the development and evaluation of the BAEA* for mobile robot path planning, based on the reference [18]. The algorithm has remarkable adaptability to dynamically changing environments and ensures robust navigation in the context of environmental changes. Its scale further enhances its applicability in large and complex environments, which means it has flexibility for various practical applications. The rigorous evaluation of our proposed BAEA* algorithm with the Bi-directional adaptive A* (BAA*) algorithm [18] in five different environments demonstrates the superiority of the BAEA* algorithm.Research limitations/implicationsThe rigorous evaluation of our proposed BAEA* algorithm with the BAA* algorithm [18] in five different environments demonstrates the superiority of the BAEA* algorithm. The BAEA* algorithm consistently outperforms BAA*, demonstrating its ability to plan shorter and more stable paths and achieve higher success rates in all environments.Originality/valueThe originality of this paper lies in the introduction of the bidirectional adaptive enhancing A* algorithm (BAEA*) as a novel solution for path planning for mobile robots. This algorithm is characterized by its unique characteristics that distinguish it from others in this field. First, BAEA* uses a unique bidirectional search strategy, allowing to explore the same path from both the initial node and the target node. This approach significantly improves efficiency by quickly converging to the best paths and using A* heuristic knowledge. In particular, the algorithm shows remarkable capabilities to quickly recognize shorter and more stable paths while ensuring higher success rates, which is an important feature for time-sensitive applications. In addition, BAEA* shows adaptability and robustness in dynamically changing environments, not only avoiding obstacles but also respecting various constraints, ensuring safe path selection. Its scale further increases its versatility by seamlessly applying it to extensive and complex environments, making it a versatile solution for a wide range of practical applications. The rigorous assessment against established algorithms such as BAA* consistently shows the superior performance of BAEA* in planning shorter paths, achieving higher success rates in different environments and cementing its importance in complex and challenging environments. This originality marks BAEA* as a pioneering contribution, increasing the efficiency, adaptability and applicability of mobile robot path planning methods.

Defining Mechanistic Pathways for Anthropogenic Noise Impact on Avian Species

Abstract Purpose of Review This review collates and analyses data on noise exposure of birds in relation to avian hearing system performance. It provides new insights into the mechanistic pathways of anthropogenic noise impact on avian species. Recent Findings Noise impacts both humans and wildlife. Birds are of conservation concern, given the recent reports of major global declines in bird populations and that one in eight bird species is threatened with extinction. Studies of noise impacts on birds have been, and continue to be, published. Whilst many of these studies report associations between noise and a response in birds, relatively few provide clear demonstration of the mechanisms of impact. Summary Anatomical and physiological datasets were compiled for species representing nineteen avian orders. Information on noise sources, propagation path and habitat selection was also collated. Bird order was not a good predictor of bird hearing frequencies, but body dimensions were. In general, smaller birds were found to have higher peak hearing frequencies than larger birds. Cranium height was the strongest predictor of peak hearing frequency for birds. These findings provide mechanistic context to noise impacts on birds and a potential basis for predicting responses of avian species to different noise environments.

Double Deep Q-Learning-Based Path Selection and Service Placement for Latency-Sensitive Beyond 5G Applications

Double Deep Q-Learning-Based Path Selection and Service Placement for Latency-Sensitive Beyond 5G Applications

Intelligent Agents and Causal Inference: Enhancing Decision-Making through Causal Reasoning

This study examines the incorporation of causal inference methods into intelligent entities and examines the benefits of utilizing causal reasoning to improve decision-making procedures. This study entails conducting an experimental evaluation within a video game setting to evaluate the performance of three separate agent types: ExplorerBOT, GuardBOT, and CausalBOT. The ExplorerBOT utilizes a stochastic path selection technique for task completion, whereas the GuardBOT remains immobile yet exhibits exceptional proficiency in identifying and neutralizing other bots. On the other hand, the CausalBOT utilizes sophisticated causal inference methods to examine the underlying factors contributing to the failures noticed in the task completion of the ExplorerBOT. The aforementioned feature allows CausalBOT to make informed decisions by selecting paths that have a greater likelihood of achieving success. The main purpose of these experiments is to assess and compare the effectiveness of two distinct bots, namely ExplorerBOT and CausalBOT, in accomplishing their respective objectives. To facilitate comparison, two iterations of the ExplorerBOT are utilized. The initial iteration is predicated exclusively on stochastic path selection and necessitates a more profound understanding of the variables that impact the achievement of tasks. On the other hand, the second version integrates an algorithm for informed search. In contrast, CausalBOT employs causal inference techniques to discover the underlying causes of failures exhibited by ExplorerBOTs and collect pertinent data. Through the process of discerning the fundamental causal mechanisms, CausalBOT is able to make well-informed decisions by selecting pathways that maximize the probability of successfully completing a given job. The utilization of this approach greatly boosts the decision-making powers of CausalBOT, hence enabling it to effectively adapt and overcome problems in a more efficient manner when compared to alternative agents.

Path Optimization of Container Multimodal Transportation Considering Differences in Cargo Time Sensitivity

To optimize the container multimodal transportation path selection problem with differences in cargo time sensitivity, this paper introduces the concept of mixed time windows and establishes an integer programming model from the perspective of carriers. The objective is to minimize the total transportation cost while meeting the customer’s on-time delivery requirements. The model imposes hard time window constraints on cargoes with on-time delivery requirements and soft time window constraints on general cargoes. A bi-level genetic algorithm is employed to solve the model. A case study is conducted to demonstrate the effectiveness of the proposed approach. Experimental results show that the bi-level genetic algorithm’s convergence ability, efficiency, stability, and global search capabilities are far superior to those of traditional single-level genetic algorithms. This indicates that the use of multi-stage solution approaches can mitigate the shortcomings of genetic algorithms in dealing with high-dimensional problems. In addition, when analyzing the effects of time penalty intensity and time window on decision-making results, it is found that the more time-sensitive cargoes are more likely to be affected by the time window setting, and will tend to sacrifice costs to ensure that the time demand is met. At the same time an increase in demand for time-sensitive services is accompanied by an increase in operating costs. Accurate assessment and classification of time sensitivity of cargoes can effectively prevent wasting resources in unnecessary areas, and ultimately maximize decision outcomes and ensure the most efficient use of resources.

A novel method for solving the multi-commodity flow problem on evolving networks

The minimum cost multi-commodity flow (MMCF) problem on evolving networks is the latest challenge to solving the min-cost network flow problem of multi-commodity flowing from multi-source and multi-sink on an evolving network. Previous researches primarily focus on the solution of MMCF problem within static networks. However, researchers ignored the MMCF problem within networks where the structure dynamically changes over time. Evolving networks have a dynamically changing and unpredictable network structure. The existing methods must recalculate when the network structure changes, leading to high computational complexity. This manuscript proposes a dynamic graph computing model for adaptive dynamic path selection and optimization. First, the model selects the path with the least cost for each commodity, assuming the route can transport a specified number of such commodities to find the optimal path faster. Second, we design path selection and flow update strategies for each node in the graph to adapt to changes in the graph structure. At the same time, we can prevent multiple commodity conflicts from exceeding the marginal capacity limit. Finally, the optimal path of each commodity is dynamically pruned based on the original optical path to obtain the tree structure and determine the absolute path to meet the specified transportation quantity of each commodity. Performance analysis results show that our method can efficiently cope with the dynamic changes of the network structure, which provides ideas for solving the multi-commodity network flow problem in dynamic networks.

A large-scale multi-attribute group decision-making method with R-numbers and its application to hydrogen fuel cell logistics path selection

AbstractThe large-scale multi-attribute group decision-making (LSMAGDM) problem has become a hot research topic in the field of decision science. An R-numbers large-scale multi-attribute group decision-making (R-LSMAGDM) model is proposed to be constructed in this paper based on the advantages of R-numbers in capturing risks. First, the most commonly used clustering method, k-means, is introduced to determine the sub-groups. Then, a new sub-group weighting determination model is constructed by considering sub-group size and sub-group entropy. Next, we also build an optimized consensus-reaching model by improving the calculation method of the mean value. Then, the R-numbers weighted Hamy mean (RNWHM) operator is proposed to aggregate the sub-group information. In addition, the logarithmic percentage change-driven objective weighting (LOPCOW) method and the compromise ranking of alternatives from distance to ideal solution (CRADIS) method are used for attribute weighting calculation and alternative ranking, respectively. Finally, the effectiveness of the model is verified by an application example of hydrogen fuel cell logistics path selection.

Research on the Influence of Strategic Adjustment on Enterprise Performance from The Perspective of Mixed Reform of State-owned Enterprises

 In order to effectively solve the problems of low operation efficiency and rigid management system of state-owned enterprises, governments at all levels have actively implemented and practiced mixed ownership reform since 2016, trying to promote industrial optimization and upgrading through mixed reform and stimulate the internal development potential of state-owned enterprises. With the continuous development of the economy and society, the state adopted the three-year Action Plan for SoE Reform (2020-2022) in June 2020, giving guiding opinions and new requirements for further deepening the reform of soes. In this context, to help has not yet been mixed change of state-owned enterprises to better understand the concrete implementation process, Shanxi Fen Wine, for example, starting from the mixed change for, the enterprise mixed change in-depth after expanding brand depth, integrated marketing resources, strengthen scientific research innovation, innovation product strategy, optimize the governance structure of five strategic adjustment direction, and the strategic adjustment of influence on enterprise performance. Finally, it draws enlightenment from the three aspects of path selection of mixed reform, advantage utilization of mixed reform and control of strategic direction, which provides reference for the strategic adjustment of mixed reform of state-owned enterprises in the industry. The research found that Shanxi Fen Wine combined with the advantages of mixed reform and made strategic adjustments around the five aspects of brand, marketing, scientific research, product and governance, which had a positive and positive impact on the financial performance of enterprises.

Universal slope-based J-integral methods for characterization of the mode I, mode II and mixed mode I/II fracture behaviour of adhesively bonded interfaces

Universal slope-based J-integral methods have been developed for the determination of the energy release rate for adhesively bonded joints under mode I, mode II and mixed-mode (I/II) loading conditions. The individual J components corresponding to the mode I and mode II loading were separated based on the J-integral decomposition theory. The proposed methods use the slopes of the substrates at various locations to characterize the energy release rate and thus avoid the measurement of crack lengths, which are especially suitable for characterizing the tough interfaces associated with large fracture process zones ahead of crack tips. Under linear elastic deformation, the slope-based J equations were found to be equivalent to classical G equations based on linear elastic fracture mechanics (LEFM). Both experimental and numerical testing of adhesively bonded joints were undertaken to validate the slope-based J equations. The universal slope-based J-integral methods provide a reliable alternative to the measurement of G for adhesive joints or laminated composites undergoing nonlinear or inelastic deformations where conventional LEFM is not valid. It is shown that LEFM, even when coupled with an effective crack length approach, can be inaccurate when damage occurs in a test specimen away from the fracture process zone, as was seen here in mode II. Slope-based J equations can avoid these inaccuracies with a careful selection of contour paths. Slope based methods are therefore strong candidates for selection in future test standards for mode II fracture characterisation of structural adhesive joints.

High-fidelity entanglement routing in quantum networks

Quantum networks, endowed with their distinct quantum properties, are poised to emerge as indispensable platforms for the implementation of quantum information technology applications in the near future. Within the realm of quantum networks, the transmission of remote quantum information presents a pivotal and intricate challenge, commonly referred to as the entanglement routing problem. While previous efforts have delved into the intricacies of entanglement routing, there has been a notable oversight concerning the crucial parameter of entanglement fidelity. Hence, this paper introduces a novel Purification-enabled Entanglement Routing Algorithm (PERA). Initially, PERA utilizes the network throughput as the routing metric and ensures fidelity guarantees on all nodes in the link through a hop-by-hop purification process. Subsequently, acknowledging the probabilistic nature of entanglement purification, a strategy involving multiple attempts at purification is employed. Finally, a comprehensive utility function is deployed to address the path selection problem for multiple source–destination pairs. Numerical simulations confirm that PERA establishes entanglement connections with outstanding fidelity guarantees. Concurrently, PERA outperforms the baseline algorithm in terms of network throughput and resource utilization.

A stochastic model of ant trail formation and maintenance in static and dynamic environments

AbstractColonies of ants can complete complex tasks without the need for centralised control as a result of interactions between individuals and their environment. Particularly remarkable is the process of path selection between the nest and food sources that is essential for successful foraging. We have designed a stochastic model of ant foraging in the absence of direct communication. The motion of ants is governed by two components - a random change in direction of motion that improves ability to explore the environment, and a non-random global indirect interaction component based on pheromone signalling. Our model couples individual-based off-lattice ant simulations with an on-lattice characterisation of the pheromone diffusion. Using numerical simulations we have tested three pheromone-based model alternatives: (1) a single pheromone laid on the way toward the food source and on the way back to the nest; (2) single pheromone laid on the way toward the food source and an internal imperfect compass to navigate toward the nest; (3) two different pheromones, each used for one direction. We have studied the model behaviour in different parameter regimes and tested the ability of our simulated ants to form trails and adapt to environmental changes. The simulated ants behaviour reproduced the behaviours observed experimentally. Furthermore we tested two biological hypotheses on the impact of the quality of the food source on the dynamics. We found that increasing pheromone deposition for the richer food sources has a larger impact on the dynamics than elevation of the ant recruitment level for the richer food sources.

Integrated optimization of scheduling for unmanned follow-me cars on airport surface

To promote the application of automated vehicles in large airports, in this study, we present an integrated optimization method for scheduling Unmanned follow-me cars. The scheduling process is divided into three phases: Dispatch, Guidance, and Recycle. For the Dispatch phase, we establish a vehicle assignment model, to allocate the vehicle resource equitably. For the Guidance phase, we offer an quantitative way, to measure the spacing between Unmanned follow-me car and aircraft. To optimize the efficiency of airport operation in the three phases and ensure safety, the collaborative planning model, and the conflict prediction model are established. An improved grey wolf optimization algorithm is adopted to enhance the convergence speed and generalization performance. A case study at Ezhou Huahu Airport in China demonstrates the effectiveness of the methods. The results show that the model of collaborative planning can make the balance of path selection, Unmanned follow-me car’s working time, and departure sequence. The convergence speed of the improved algorithm has been increased by 18.75%. The inequity index of vehicle assignment is only 0.015731, and the spatiotemporal distribution of conflicts is influenced by the airport’s surface layout.

Memory-Enhanced Knowledge Reasoning with Reinforcement Learning

In recent years, the emergence of large-scale language models, such as ChatGPT, has presented significant challenges to research on knowledge graphs and knowledge-based reasoning. As a result, the direction of research on knowledge reasoning has shifted. Two critical issues in knowledge reasoning research are the algorithm of the model itself and the selection of paths. Most studies utilize LSTM as the path encoder and memory module. However, when processing long sequence data, LSTM models may encounter the problem of long-term dependencies, where memory units of the model may decay gradually with an increase in time steps, leading to forgetting earlier input information. This can result in a decline in the performance of the LSTM model in long sequence data. Additionally, as the data volume and network depth increase, there is a risk of gradient disappearance. This study improved and optimized the LSTM model to effectively address the problems of gradient explosion and gradient disappearance. An attention layer was employed to alleviate the issue of long-term dependencies, and ConvR embedding was used to guide path selection and action pruning in the reinforcement learning inference model. The overall model achieved excellent reasoning results.

Some necessary constructs of consciousness as defined in the dynamic organicity theory

The Dynamic Organicity Theory (DOT) explains consciousness as a process that cannot be reduced to a thing or a structure. This process, called the Polanyian process, involves the evolving diachronic boundary conditions of our experienceability. These boundary conditions are ontologically intertwined with higher-level boundary conditions, which lead to physiological nonlocal cause-and-effect relationships. Thus, adaptive changes occur when self-referential causal closure transforms syntactic structures into experienceable forms. Self-referential causal closure is a causal agent for diachronic boundary conditions that majorly structure intrinsic information in ways that experienceabilities are across time, suggesting that consciousness is not driven by an “internal clock” or regular biological rhythm but rather by intrinsic intentionality preceding path selection through a non-mechanical force of hidden thermodynamic energy that cascades to form a volitional agency of information-based action due to the gain of hidden intrinsic information. The process of transforming syntactical structures to experienceable forms in intrinsic informational pathways is conflated with nonlinear time, and its structuring defines the functionality of the brain. Consciousness-in-the-moment is an “averaging out” between different informational pathways involved across such nonlinear time. The transduction of quantum potential information as negentropic action is a motif for “potential complexity” to decrease the maximum complexity (derived from the brain structure, dynamics, and function) due to the reduction of uncertainty. Given that the functionality of maximum complexity lacks the self-referential causal closure to be a volitional agent needed for precognitive consciousness, it is proposed that negentropic entanglement unifies experienceabilities so that the functionality of multiscale complexity is greater than the functionality of maximum complexity. This may be enough to produce consciousness in organic systems.

An Agile Super-Resolution Network via Intelligent Path Selection

In edge computing environments, limited storage and computational resources pose significant challenges to complex super-resolution network models. To address these challenges, we propose an agile super-resolution network via intelligent path selection (ASRN) that utilizes a policy network for dynamic path selection, thereby optimizing the inference process of super-resolution network models. Its primary objective is to substantially reduce the computational burden while maximally maintaining the super-resolution quality. To achieve this goal, a unique reward function is proposed to guide the policy network towards identifying optimal policies. The proposed ASRN not only streamlines the inference process but also significantly boosts inference speed on edge devices without compromising the quality of super-resolution images. Extensive experiments across multiple datasets confirm ASRN’s remarkable ability to accelerate inference speeds while maintaining minimal performance degradation. Additionally, we explore the broad applicability and practical value of ASRN in various edge computing scenarios, indicating its widespread potential in this rapidly evolving domain.

TEE-AODV Trust-based Route Selection and Improving Energy Efficiency in MANET

Mobile ad hoc networks (MANETs) are the future communication infrastructure for all devices that operate without the elaborate design found in wired infrastructure networks. Constructing a routing protocol that helps MANETs to achieve (QoS). It is quite challenging to achieve energy-efficient routing, due to its limited resources and node mobility concerns, which would impact node resource stability and lead to congestion and lower user QoS. A novel Trustworthy Energy Efficient AODV protocol has been proposed for improving path selection and locating trusted stations to maximize QoS. This approach employs location data, transmission speed, and directions to estimate the energy usage and trustworthiness of distinct nodes. The proposed TEE-AODV protocol includes two new control information namely Trust Request packet and Trust Reply. (T-RREP) AODV's packet includes a neighbor list and channel trustworthiness. The number of working neighbors in every path is counted, and this information is used to efficiently choose paths. The TEE-AODV calculates a QoS supporting metric for each detected route to identify a traffic-free path at any moment and determines whether it is a trusted route. The NS-2 simulator has been used to assess the suggested approach in terms of specific characteristics, including energy consumption (EC), (PDR), end-to-end delay (), and number of dead nodes. According to the experimental results, the suggested framework enhances reliable route selection while lowering e nd-to-end latency and improving energy efficiency. The experimental findings indicate that the suggested TEE-AODV protocol achieves high energy efficiency rate in the selection of trustworthy route and minimizing the E D .

Study of the Point Method for fretting fatigue life prediction

With the improvement and success of the Theory of Critical Distances (TCD) in notch fatigue problems, the TCD, especially the Point Method (PM), combined with multiaxial fatigue criteria, has been used over the last few years to predict the fretting fatigue life. However, little attention is paid to the effect of different damage mechanisms, especially stress distribution, between the two conditions on the application of the PM. This paper aims to select a suitable path for the PM, considering various stress distributions, for the fretting fatigue life prediction of complete contacts. A new method is developed to select the suitable path of the PM for fretting fatigue life prediction. To validate the analysis, corresponding tests were carried out using an improved proving ring-like rig in a Ti-6Al-4V alloy. Predicted lives on all paths were calculated using the PM combined with different multiaxial fatigue criteria, without incorporating wear effects, and compared with experimental lives using an error bar graph. Results indicate that when the fretting wear isn't considered, the minimum predicted life path is more suitable for the PM than the perpendicular and maximum predicted life paths. This suitable path is compared to the path selection in the uniaxial symmetric V-notch fatigue condition. Analysis shows that the conventional path (the notch bisector) of the PM produces the maximum predicted life for the notch fatigue condition. The cause of this different path selection of the PM in the two fatigue conditions is assumed to be the effect of the asymmetric stress distribution and fretting wear damage in the fretting fatigue condition.

Interaction of atmospheric pressure helium plasma jet with non-planar substrates: path selectivity of surface ionization wave

Most surfaces treated by atmospheric pressure plasma jets (APPJs) in practical applications are notably three-dimensional. However, non-planar surfaces exhibit a diverse array of geometries, such as variations in curvature, roughness, and texture, complicating the prediction of surface ionization waves (SIWs) propagation behavior across varied surface shapes, in the absence of sufficient experimental data. In this study, we made measurements of APPJ interactions with the non-planar substrates using the spatio-temporal resolved image method. Non-planar substrates encompassed wavy surfaces, arrayed hemispheres, and randomly textured raised surfaces. We tracked the morphology and velocity of SIW propagation over these surfaces. The results indicate that the SIW propagation on non-planar surfaces is significantly influenced by surface geometry and displays path selectivity, i.e. the SIW tends to propagate along valleys. The average propagation velocity of the SIW increases with the increasing radius of the wavy surface, as well as with the increased height of the arrayed hemispheres. This is attributable to the surface geometry constraining the dispersion of the SIW, causing it to concentrate and propagate in a singular direction. Moreover, the surface geometry markedly affects the distribution of the plasma treatment area, with the plasma inclined to enter valleys (where the light emission is significantly stronger than that of peaks) and to closely adhere to hemispherical surfaces. These patterns suggest a potential positive impact on treating skin surfaces such as pores, reducing bacteria in wrinkles, and addressing pimples.

A 0.5–15 GHz reconfigurable IR-UWB receiver in 28 nm CMOS

A 0.5–15 GHz ultra-wideband (UWB) RF reconfigurable receiver in UMC 28 nm CMOS used for IR-UWB (impulse radio UWB) radar system is proposed in this paper. This receiver employs a 0.5–8 GHz path and an 8–15 GHz path to achieve the wide frequency band, and we proposed an RF channel select voltage-to-time converter (VTC) circuit. Two different low noise amplifiers (LNAs) are adopted in the two paths to obtain the high performance of noise figure (NF) and gain. The common drain stage and resistance negative feedback technology as well as the source degeneration inductive technology are introduced to achieve a high flat gain and fine input matching. The R-2R variable load network of a programable gain amplifier (PGA) gives it a nearly 24 dB dynamic range and approximately linear gain control. A 500MS/s time domain analog to digital converter (TD-ADC) with an equivalent sampling rate of 16GS/s is shared for the two paths which can realize the direct RF sampling. The input RF signal path selection is implemented by configuring the Band_sel signal. The receiver has reconfigurability of the RF front end and better power consumption and area. In the frequency of 0.5–8 GHz, the S11 is below −7.9 dB and the NF is smaller than 6 dB. In the frequency of 8–15 GHz, the S11 is below −8dB and the NF is smaller than 11 dB. The receiver dissipates 242.72mW with an area of 1.01 mm2

Intelligent Routing in Directional Ad Hoc Networks Through Predictive Directional Heat Map From Spatio-Temporal Deep Learning

By applying a simple shortest/minimum-cost routing algorithm, the mobile ad-hoc network (MANET) with heavy data transmissions may be easily congested if multiple routing paths meet at the same relay node. A path may be unable to provide high throughput since some of its path nodes are helping other paths to forward the heavy traffic. Therefore, those busy nodes should be avoided when a new path is to be established. The task of optimal path seeking becomes more challenging when a MANET is equipped with directional antennas that may cause directional interference with neighboring receivers. The motivation of our research is to build an intelligent proactive routing scheme for MANETs with directional antennas. Our directional routing protocol considers not only the global traffic distribution in different areas of the MANET, but also the properties of directional antennas. It uses spatio-temporal deep learning algorithm to predict the next-time snapshot of directional heat map (DHM), which shows the traffic density distribution in each network location as well as the coverage of each directional antenna. Such a DHM will be used to identify the optimal path that can avoid congested areas as well as the interference from all neighboring directional links. Furthermore, an optimization algorithm is designed to perform optimal path selection. It is able to automatically split a single path into multiple ones that later on get converged again into one path, if the path needs to go around a congested area. Therefore, our routing scheme achieves better quality-of-service (QoS) performance than traditional routing mechanisms such as AODV and OLSR.