Abstract While machine learning's role in financial trading has advanced considerably, algorithmic transparency and explainability challenges still exist. This research enriches prior studies focused on high‐frequency financial data prediction by introducing an explainable reinforcement learning model for portfolio management. This model transcends basic asset prediction, formulating concrete, actionable trading strategies. The methodology is applied in a custom trading environment mimicking the CAC‐40 index's financial conditions, allowing the model to adapt dynamically to market changes based on iterative learning from historical data. Empirical findings reveal that the model outperforms an equally weighted portfolio in out‐of‐sample tests. The study offers a dual contribution: it elevates algorithmic planning while significantly boosting transparency and interpretability in financial machine learning. This approach tackles the enduring ‘black‐box’ issue and provides a holistic, transparent framework for managing investment portfolios.

This paper introduces a novel hybridisation technique combining the Backtracking Search (BS) and Differential Evolution (DE) algorithms. The proposed hybridisation executes diversity loss and stagnation detection mechanisms to maintain the diversity of the populations, in addition, modifications are done over the mutation operators of the component algorithms in order to improve the search capability of the proposal. These modifications are self-adapted and implemented simultaneously. Extensive experiments to establish the optimal configuration of the parameters are also presented through the introduced technique. The proposed hybridisation approach has been applied to five classical versions and two state-of-the-art variants of DE and tested against 28 well-known benchmark functions with different dimensions, each type of which highlights a different set of characteristics and provides a baseline measurement to validate the performance of the algorithms. In order to further test the proposal, the four outstanding algorithms in the state of the art have also been included in the comparisons. Experimental results show the effectiveness of the proposed hybrid framework over the compared algorithms.

Classifier ensemble pruning is a strategy through which a subensemble can be identified via optimizing a predefined performance criterion. Choosing the optimum or suboptimum subensemble decreases the initial ensemble size and increases its predictive performance. In this article, a set of heuristic metrics will be analyzed to guide the pruning process. The analyzed metrics are based on modifying the order of the classifiers in the bagging algorithm, with selecting the first set in the queue. Some of these criteria include general accuracy, the complementarity of decisions, ensemble diversity, the margin of samples, minimum redundancy, discriminant classifiers, and margin hybrid diversity. The efficacy of those metrics is affected by the original ensemble size, the required subensemble size, the kind of individual classifiers, and the number of classes. While the efficiency is measured in terms of the computational cost and the memory space requirements. The performance of those metrics is assessed over fifteen binary and fifteen multiclass benchmark classification tasks, respectively. In addition, the behavior of those metrics against randomness is measured in terms of the distribution of their accuracy around the median. Results show that ordered aggregation is an efficient strategy to generate subensembles that improve both predictive performance as well as computational and memory complexities of the whole bagging ensemble.

The knowledge of the number of tags is critical in many radio frequency identification (RFID) applications. This paper is concerned with the problem of estimating an RFID tag population size when the number of tags is much higher than the frame size. A novel estimation scheme called “Scalable Minimum Mean Square Error” (sMMSE) is proposed. The proposed estimator updates the frame size by considering two key parameters: One determines the limit of the slots occupancy at which the frame size should be increased, and the other one sets the frame size increase factor. A formal study is provided to adjust these parameters with the aim of lowering the estimation error while scaling to highly populated tag sets. Numerical results indicate that sMMSE significantly decreases the normalized estimation error and maintains a low estimation time compared to existing strategies.

An improved solution for drug distribution is presented in this paper. It is divided into two parts: i) a multi-operator evolutionary algorithm in charge of calculating the initial delivery routes and ii) an ambient intelligence-based support system able to tracing the merchandise along the distribution route. The first one establishes the routes to be followed by the vehicles, based on a proposal of estimation of the travel times. The second one is formed by a system able to recognize and trace the drugs inside each vehicle. A laboratory experimentation has been conducted in order to demonstrate the adequacy of the route calculator. In addition, a field experimentation has been carried out by implementing the traceability system in a delivery van which of the drug distributor in the city of Bilbao.

This paper presents a method of optimizing the elements of a hierarchy of fuzzy-rule-based systems (FRBSs). It is a hybridization of a genetic algorithm (GA) and the cross-entropy (CE) method, which is here called GACE. It is used to predict congestion in a 9-km-long stretch of the I5 freeway in California, with time horizons of 5, 15, and 30 min. A comparative study of different levels of hybridization in GACE is made. These range from a pure GA to a pure CE, passing through different weights for each of the combined techniques. The results prove that GACE is more accurate than GA or CE alone for predicting short-term traffic congestion.

Up to date energy-aware radio frequency identification (RFID) anticollision protocols are mainly focused on RFID systems using active tags. The introduction of RFID in portable devices, and the increasing popularity of passive tags, demands low energy-consumption anticollision protocols in environments with passive tags. Memoryless protocols present the simplest tag hardware that is beneficial for a low energy consumption. However, the excessive collisions required by these protocols increases the energy consumption, wasting a large number of tag transmitting bits. In this paper, a window procedure is presented to manage the length of the tags' bit-stream in order to limit the energy wastage in collisions. The bit window is applied to two memoryless anticollision protocols, the query tree protocol (QT) and the collision tree protocol (CT), producing the query window tree protocol (QwT) and the collision window tree protocol (CwT). Simulations show that the use of this window yields significant reductions in energy consumption and identification time. In addition, QwT and CwT are compared to state of the art memoryless protocols to confirm that CwT is an outstanding candidate for reducing energy consumption. Also, the results obtained show how the bit window procedure increases energy savings as the length of the tag ID increases.

Nowadays, some developments in the vehicle industry permit a safe and comfortable driving. However, several manufactures and research groups are still working in the improvement of the control strategies and path smoothing algorithms. In this paper, a new trajectory generation approach for autonomous vehicles in urban scenarios, considering parametric equations, is proposed. An algorithm that considers Bezier curves and circumference parametric equations for a real vehicle, specifically in roundabout and urban intersections is presented. This approach is generated in real time and can be adapted to dynamic changes in the route. A smooth trajectory generator computationally efficient and easily implementable is proposed. Moreover, this new trajectory generator reduces the control actions, generated with to a fuzzy controller. Some trials have been performed in an urban circuit with promising performance.

We propose a multi-crossover and adaptive island based population algorithm (MAIPA). This technique divides the entire population into subpopulations, or demes, each with a different crossover function, which can be switched according to the efficiency. In addition, MAIPA reverses the philosophy of conventional genetic algorithms. It gives priority to the autonomous improvement of the individuals (at the mutation phase), and introduces dynamism in the crossover probability. Each subpopulation begins with a very low value of crossover probability, and then varies with the change of the current generation number and the search performance on recent generations. This mechanism helps prevent premature convergence. In this research, the effectiveness of this technique is tested using three well-known routing problems, i.e., the traveling salesman problem (TSP), capacitated vehicle routing problem (CVRP), and vehicle routing problem with backhauls (VRPB). MAIPA proves to be better than a traditional island based genetic algorithm for all these three problems.

Autonomous navigation is one of the most important challenges in the outdoor mobile robot field. For an automatic vehicle (which can be considered a type of outdoor mobile robot), path following can be implemented using global positioning systems (GPS) to allow the configuration of different navigation styles such as the shortest or fastest route, toll avoidance, etc., and even the definition of new routes. The main problem is when an unexpected circumstance occurs - traffic accident, road closure, etc. This paper presents an autonomous vehicle guidance system based on fuzzy logic systems to resolve unexpected road situations. A fuzzy steering controller performs the autonomous navigation, allowing reverse as well as forward driving in urban environments. Good performance was obtained in trials performed with a commercial electric Citroën Berlingo van on a private driving circuit.

This work presents a driving system designed for virtual racing situations. It is based on a complete modular architecture capable of automatically driving a car along a track with or without opponents. The architecture is composed of intuitive modules, with each one being responsible for a basic aspect of car driving. Moreover, this modularity of the architecture will allow us to replace or add modules in the future as a way to enhance particular features of particular situations. In the present work, some of the modules are implemented by means of hand-designed driving heuristics, whereas modules responsible for adapting the speed and direction of the vehicle to the track's shape, both critical aspects of driving a vehicle, are optimized by means of a genetic algorithm that evaluates the performance of the controller in four different tracks to obtain the best controller in a large number of situations; the algorithm also penalizes controllers that go out of the track, lose control, or get damaged. The evaluation of the performance is done in two ways. First, in runs with and without adversaries over several tracks. And second, the architecture was submitted as a participant to the 2010 Simulated Car Racing Competition, which in end won laurels. © 2012 Wiley Periodicals, Inc.

Vehicles equipped with intelligent systems designed to prevent accidents, such as collision warning systems (CWSs) or lane-keeping assistance (LKA), are now on the market. The next step in reducing road accidents is to coordinate such vehicles in advance not only to avoid collisions but to improve traffic flow as well. To this end, vehicle-to-infrastructure (V2I) communications are essential to properly manage traffic situations. This paper describes the AUTOPIA approach toward an intelligent traffic management system based on V2I communications. A fuzzy-based control algorithm that takes into account each vehicle's safe and comfortable distance and speed adjustment for collision avoidance and better traffic flow has been developed. The proposed solution was validated by an IEEE-802.11p-based communications study. The entire system showed good performance in testing in real-world scenarios, first by computer simulation and then with real vehicles.

In this work, a solution for clustering and tracking obstacles in the area covered by a LIDAR sensor is presented. It is based on a combination of simple artificial intelligence techniques and it is conceived as an initial version of a detection and tracking system for objects of any shape that an autonomous vehicle might find in its surroundings. The proposed solution divides the problem into three consecutive phases: 1) segmentation, 2) fragmentation detection and clustering and 3) tracking. The work done has been tested with real world LIDAR scan samples taken from an instrumented vehicle.

Research on intelligent transport systems (ITSs) is steadily leading to safer and more comfortable control for vehicles. Systems that permit longitudinal control have already been implemented in commercial vehicles, acting on throttle and brake. Nevertheless, lateral control applications are less common in the market. Since a too-sudden turn of the steering wheel can cause an accident in a few seconds, good speed and position control of the steering wheel is essential. We present here a new cascade control architecture based on fuzzy logic controllers that emulate a human driver's behavior. The control architecture was tested on a real vehicle at different vehicle speeds. The results showed the use of a straightforward and intuitive fuzzy controller to give good performance.

The control of nonlinear systems has been putting especial attention in the use of Artificial Intelligent techniques, where fuzzy logic presents one of the best alternatives due to the exploit of human knowledge. However, several fuzzy logic real-world applications use manual tuning (human expertise) to adjust control systems. On the other hand, in the Intelligent Transport Systems (ITS) field, the longitudinal control (throttle and brake management) is an important topic because external perturbations can generate uncomfortable accelerations as well as unnecessary fuel consumption. In this work, we utilize a neuro-fuzzy system to use human driving knowledge to tune and adjust the input-output parameters of a fuzzy if-then system. The neuro-fuzzy system considered in this work is ANFIS (Adaptive-Network-based Fuzzy Inference System). Results show several improvements in the control system adjusted by neuro-fuzzy techniques in comparison to the previous manual tuned controller, mainly in comfort and efficient use of actuators.

In this paper, we overview the 2009 Simulated Car Racing Championship-an event comprising three competitions held in association with the 2009 IEEE Congress on Evolutionary Computation (CEC), the 2009 ACM Genetic and Evolutionary Computation Conference (GECCO), and the 2009 IEEE Symposium on Computational Intelligence and Games (CIG). First, we describe the competition regulations and the software framework. Then, the five best teams describe the methods of computational intelligence they used to develop their drivers and the lessons they learned from the participation in the championship. The organizers provide short summaries of the other competitors. Finally, we summarize the championship results, followed by a discussion about what the organizers learned about 1) the development of high-performing car racing controllers and 2) the organization of scientific competitions.

A major research topic in intelligent transportation systems (ITSs) is the development of systems that will be capable of controlling the flow of vehicular traffic through crossroads, particularly in urban environments. This could significantly reduce traffic jams, since autonomous vehicles would be capable of calculating the optimal speed to maximize the number of cars driving through the intersection. We describe the use of vehicle-to-vehicle (V2V) communications to determine the position and speed of the vehicles in an environment around a crossroad. These data are used to estimate the intersection point, and a fuzzy controller then modifies the speed of the cars without right of way according to the speed of the car with right of way. Experimental tests conducted with two mass-produced cars on a real circuit at the facilities of the Instituto de Automa¿tica Industrial, Consejo Superior de Investigaciones Cienti¿ficas, Madrid, Spain, gave excellent results.

The techniques of soft computing are recognized as having a strong learning and cognition capability as well as good tolerance to uncertainty and imprecision. These properties allow them to be applied successfully to intelligent transportation systems (ITS), a broad range of diverse technologies that designed to answer many transportation problems. The unmanned control of the steering wheel is one of the most important challenges faced by researchers in this area. This paper presents a method of automatically adjusting a fuzzy controller to manage the steering wheel of a mass-produced vehicle. Information about the state of the car while a human driver is handling it is captured and used to search, via genetic algorithms, for the best fit of an appropriate fuzzy controller. Evaluation of the fuzzy controller will take into account its adjustment to the human driver's actions and the absence of abrupt changes in its control surface, so that not only is the route tracking good, but the drive is smooth and comfortable for the vehicle's occupants.