Reinforcement Learning from Demonstrations (RLfD) has emerged as an effective method by fusing expert demonstrations into Reinforcement Learning (RL) training, harnessing the strengths of both Imitation Learning (IL) and RL. However, existing algorithms rely on offline demonstrations, which can introduce a distribution gap between the demonstrations and the actual training environment, limiting their performance. In this paper, we propose a novel approach, Reinforcement Learning from Online Demonstrations (RLfOLD), that leverages online demonstrations to address this limitation, ensuring the agent learns from relevant and up-to-date scenarios, thus effectively bridging the distribution gap. Unlike conventional policy networks used in typical actor-critic algorithms, RLfOLD introduces a policy network that outputs two standard deviations: one for exploration and the other for IL training. This novel design allows the agent to adapt to varying levels of uncertainty inherent in both RL and IL. Furthermore, we introduce an exploration process guided by an online expert, incorporating an uncertainty-based technique. Our experiments on the CARLA NoCrash benchmark demonstrate the effectiveness and efficiency of RLfOLD. Notably, even with a significantly smaller encoder and a single camera setup, RLfOLD surpasses state-of-the-art methods in this evaluation. These results, achieved with limited resources, highlight RLfOLD as a highly promising solution for real-world applications.

Palm oil production has been identified as one of the major drivers of deforestation for tropical countries. To meet supply chain objectives, commodity producers and other stakeholders need timely information of land cover dynamics in their supply shed. However, such data are difficult to obtain from suppliers who may lack digital geographic representations of their supply sheds and production locations. Here we present a "community model," a machine learning model trained on pooled data sourced from many different stakeholders, to develop a specific land cover probability map, in this case a semi-global oil palm map. An advantage of this method is the inclusion of varied inputs, the ability to easily update the model as new training data becomes available and run the model on any year that input imagery is available. Inclusion of diverse data sources into one probability map can help establish a shared understanding across stakeholders on the presence and absence of a land cover or commodity (in this case oil palm). The model predictors are annual composites built from publicly available satellite imagery provided by Sentinel-1, Sentinel-2, and ALOS DSM. We provide map outputs as the probability of palm in a given pixel, to reflect the uncertainty of the underlying state (palm or not palm). The initial version of this model provides global accuracy estimated to be approximately 90 partitioned test data. This model, and resulting oil palm probability map products are useful for accurately identifying the geographic footprint of palm cultivation. Used in conjunction with timely deforestation information, this palm model is useful for understanding the risk of continued oil palm plantation expansion in sensitive forest areas.

The development of Autonomous Driving (AD) systems in simulated environments like CARLA is crucial for advancing real-world automotive technologies. To drive innovation, CARLA introduced Leaderboard 2.0, significantly more challenging than its predecessor. However, current AD methods have struggled to achieve satisfactory outcomes due to a lack of sufficient ground truth data. Human driving logs provided by CARLA are insufficient, and previously successful expert agents like Autopilot and Roach, used for collecting datasets, have seen reduced effectiveness under these more demanding conditions. To overcome these data limitations, we introduce PRIBOOT, an expert agent that leverages limited human logs with privileged information. We have developed a novel BEV representation specifically tailored to meet the demands of this new benchmark and processed it as an RGB image to facilitate the application of transfer learning techniques, instead of using a set of masks. Additionally, we propose the Infraction Rate Score (IRS), a new evaluation metric designed to provide a more balanced assessment of driving performance over extended routes. PRIBOOT is the first model to achieve a Route Completion (RC) of 75 and 45 datasets, potentially solving the data availability issues that have hindered progress in this benchmark.

BACKGROUND:The influence of exhaustive whole-body exercise and caffeine ingestion on electromechanical delay (EMD) has been underexplored. This study investigated the effect of exhaustive cycling exercise on EMD and other parameters of muscle contractile properties and the potential ability of caffeine to attenuate the exercise-induced impairments in EMD and muscle contractile properties. METHODS:Ten healthy men cycled until exhaustion (88±2% of V̇O2max) on two separate days after ingesting caffeine (5 mg.kg-1 of body mass) or cellulose (placebo). Parameters of muscle contractile properties of the quadriceps muscles were assessed via volitional and electrically evoked isometric contractions, performed before and 50 minutes after ingestion of the capsules, and after exercise. Muscle recruitment during volitional contractions was determined via surface electromyography. RESULTS:Exhaustive cycling exercise did not affect volitional and relaxation EMD (P>0.05) but increased evoked EMD. In addition, the exhaustive cycling exercise also increased muscle recruitment at the beginning of volitional isometric muscle contraction (P<0.05). The peak twitch force, maximal rate of twitch force development, and twitch contraction time were all compromised after exhaustive cycling exercise (P<0.05). Acute caffeine ingestion had no effect on muscle contractile properties (P>0.05), except that caffeine increased twitch contraction time at postexercise (P<0.05). CONCLUSIONS:Exercise-induced decline in peripheral components of the EMD might be compensated by an increase in the muscle recruitment. In addition, acute caffeine ingestion had minimal influence on exercise-induced changes in muscle contractile proprieties.

Current approaches of Reinforcement Learning (RL) applied in urban Autonomous Driving (AD) focus on decoupling the perception training from the driving policy training. The main reason is to avoid training a convolution encoder alongside a policy network, which is known to have issues related to sample efficiency, degenerated feature representations, and catastrophic self-overfitting. However, this paradigm can lead to representations of the environment that are not aligned with the downstream task, which may result in suboptimal performances. To address this limitation, this paper proposes RLAD, the first Reinforcement Learning from Pixels (RLfP) method applied in the urban AD domain. We propose several techniques to enhance the performance of an RLfP algorithm in this domain, including: i) an image encoder that leverages both image augmentations and Adaptive Local Signal Mixing (A-LIX) layers; ii) WayConv1D, which is a waypoint encoder that harnesses the 2D geometrical information of the waypoints using 1D convolutions; and iii) an auxiliary loss to increase the significance of the traffic lights in the latent representation of the environment. Experimental results show that RLAD significantly outperforms all state-of-the-art RLfP methods on the NoCrash benchmark. We also present an infraction analysis on the NoCrash-regular benchmark, which indicates that RLAD performs better than all other methods in terms of both collision rate and red light infractions.

Texture mapping can be defined as the colorization of a 3D mesh using one or multiple images. In the case of multiple images, this process often results in textured meshes with unappealing visual artifacts, known as texture seams, caused by the lack of color similarity between the images. The main goal of this work is to create textured meshes free of texture seams by color correcting all the images used. We propose a novel color-correction approach, called sequential pairwise color correction, capable of color correcting multiple images from the same scene, using a pairwise-based method. This approach consists of sequentially color correcting each image of the set with respect to a reference image, following color-correction paths computed from a weighted graph. The color-correction algorithm is integrated with a texture-mapping pipeline that receives uncorrected images, a 3D mesh, and point clouds as inputs, producing color-corrected images and a textured mesh as outputs. Results show that the proposed approach outperforms several state-of-the-art color-correction algorithms, both in qualitative and quantitative evaluations. The approach eliminates most texture seams, significantly increasing the visual quality of the textured meshes.

Collecting real-world data is often considered the bottleneck of Artificial Intelligence, stalling the research progress in several fields, one of which is camera localization. End-to-end camera localization methods are still outperformed by traditional methods, and we argue that the inconsistencies associated with the data collection techniques are restraining the potential of end-to-end methods. Inspired by the recent data-centric paradigm, we propose a framework that synthesizes large localization datasets based on realistic 3D reconstructions of the real world. Our framework, termed Synfeal: Synthetic from Real, is an open-source, data-driven simulator that synthesizes RGB images by moving a virtual camera through a realistic 3D textured mesh, while collecting the corresponding ground-truth camera poses. The results validate that the training of camera localization algorithms on datasets generated by Synfeal leads to better results when compared to datasets generated by state-of-the-art methods. Using Synfeal, we conducted the first analysis of the relationship between the size of the dataset and the performance of camera localization algorithms. Results show that the performance significantly increases with the dataset size. Our results also suggest that when a large localization dataset with high quality is available, training from scratch leads to better performances. Synfeal is publicly available at https://github.com/DanielCoelho112/synfeal.

Texture mapping of 3D models using multiple images often results in textured meshes with unappealing visual artifacts known as texture seams. These artifacts can be more or less visible, depending on the color similarity between the used images. The main goal of this work is to produce textured meshes free of texture seams through a process of color correcting all images of the scene. To accomplish this goal, we propose two contributions to the state-of-the-art of color correction: a pairwise-based methodology, capable of color correcting multiple images from the same scene; the application of 3D information from the scene, namely meshes and point clouds, to build a filtering procedure, in order to produce a more reliable spatial registration between images, thereby increasing the robustness of the color correction procedure. We also present a texture mapping pipeline that receives uncorrected images, an untextured mesh, and point clouds as inputs, producing a final textured mesh and color corrected images as output. Results include a comparison with four other color correction approaches. These show that the proposed approach outperforms all others, both in qualitative and quantitative metrics. The proposed approach enhances the visual quality of textured meshes by eliminating most of the texture seams.

Autonomous driving in urban environments requires intelligent systems that are able to deal with complex and unpredictable scenarios. Traditional modular approaches focus on dividing the driving task into standard modules, and then use rule-based methods to connect those different modules. As such, these approaches require a significant effort to design architectures that combine all system components, and are often prone to error propagation throughout the pipeline. Recently, end-to-end autonomous driving systems have formulated the autonomous driving problem as an end-to-end learning process, with the goal of developing a policy that transforms sensory data into vehicle control commands. Despite promising results, the majority of end-to-end works in autonomous driving focus on simple driving tasks, such as lane-following, which do not fully capture the intricacies of driving in urban environments. The main contribution of this paper is to provide a detailed comparison between end-to-end autonomous driving systems that tackle urban environments. This analysis comprises two stages: a) a description of the main characteristics of the successful end-to-end approaches in urban environments; b) a quantitative comparison based on two CARLA simulator benchmarks (CoRL2017 and NoCrash). Beyond providing a detailed overview of the existent approaches, we conclude this work with the most promising aspects of end-to-end autonomous driving approaches suitable for urban environments.

Sheet metal forming tools, like stamping presses, play an ubiquitous role in the manufacture of several products. With increasing requirements of quality and efficiency, ensuring maximum uptime of these tools is fundamental to marketplace competitiveness. Using anomaly detection and predictive maintenance techniques, it is possible to develop lower risk and more intelligent approaches to maintenance scheduling, however, industrial implementations of these methods remain scarce due to the difficulties of obtaining acceptable results in real-world scenarios, making applications of such techniques in stamping processes seldom found. In this work, we propose a combination of two distinct approaches: (a) time segmentation together with feature dimension reduction and anomaly detection; and (b) machine learning classification algorithms, for effective downtime prediction. The approach (a)+(b) allows for an improvement rate up to 22.971% of the macro F1-score, when compared to sole approach (b). A ROC AUC index of 96% is attained by using Randomized Decision Trees, being the best classifier of twelve tested. An use case with a decentralized predictive maintenance architecture for the downtime forecasting of a stamping press, which is a critical machine in the manufacturing facilities of Bosch Thermo Technology, is discussed.

Table .All patients had objective improvement in gastric emptying by scintigraphic emptying studies.Patients' STC-related symptoms subjectively improved with DLI.Conclusion: Patients with combined medically refractory GP and STC are complex and require a multidisciplinary team-based approach.Addressing both GP via POP and performing a DLI may improve symptoms of both motility disorders and may better predict those who may benefit from a more invasive large bowel resection.

The development of new strategies to attenuate exercise-induced muscle damage may be helpful for training regimens. The aim of this study was to determine whether a oral formulation of angiotensin Ang-(1-7)[HPβCD/Ang-(1-7)] is effective to reduce pain, and muscle damage markers after eccentric-overload exercise. HPβCD (Placebo) and HPβCD/Ang-(1-7) (Ang-(1-7) group were treated for 7 days (one capsule/day). The pain was measured by visual analogue scale, maximal strength (MS) using force platform. Blood samples were collected for cytokines and creatine kinase (CK) analysis. The Ang-(1-7)-treated group reported less pain immediately (3.46±0.64 vs. placebo 3.80±0.77 cm) and 24 h after exercise (3.07±0.71 vs. 3.73±0.58 cm placebo) and higher MS at 24 h (24±12 N) and 48 h (30±15 N) vs. placebo (-8±9 N and -10±9 N). The CK for Ang-(1-7) (0.5±0.1 and 0.9±0.2 U/L) were lower at 48 and 72 h vs. placebo (fold changes of 1.7±0.5 and 1.5±0.3 U/L). The TNF-α level was lower in the treated group post-exercise (38±2.5 pg/ml) vs. placebo (45±2.9 pg/ml) but no significant changes were observed for IL-6 and IL-10. Our data indicate that treatment with Ang-(1-7) may attenuate pain, some of the muscle damage markers and improves performance following eccentric exercise.

Cardiorespiratory responses induced by school physical education

BACKGROUND:The aim of the present study was to evaluate the effects of T-Shirt fabric and color on the 10 km outdoor running performance.METHODS:Six men and six women (mean±SD: age: 27±5 years; height: 1.70±0.13 m; weight: 64.0±12.7 kg; body surface area: 1.73±0.29 m2; Σskinfolds: 107±24 mm; VO2max: 40.2±8.4 mL.kg-1.min-1) took part in five experimental trials, during each of which they wore: 1) no T-shirt (CON); 2) white polyester T-shirt (WP); 3) black polyester T-shirt (BP); 4) white cotton T-shirt (WC); and 5) black cotton T-shirt (BC). Average running velocity (pace) was calculated from each 2 km running time. Rectal, skin and T-shirt temperatures, heart rates and Physiological Strain Index (PSI) were measured before and after the 10 km runs and at the end of each 2 km.RESULTS:There were no differences in pace, heart rate, rectal and skin temperatures among conditions (P>0.05). PSI was higher in BC and WC conditions when compared to BP and WP conditions. T-shirt temperature was higher for the BC when compared to WP, BP and WC conditions. Rectal temperature and heart rate increased simultaneously with reduced pace throughout self-paced running (P<0.05).CONCLUSIONS:Despite fabric type T-shirt altered PSI, running performance in the 10 km run was not affected by T-shirt type or color.

BACKGROUNDPatients with idiopathic scoliosis are exposed to approximately 25 radiographic examinations of their spine throughout the clinical follow-up using the Cobb angle. Several non-invasive and radiation-free methods have been proposed to measure scoliotic deformities, including the scoliometer.OBJECTIVESTo measure the intra- and interrater reliability of the scoliometer measurements, to assess the correlation of the values obtained by the scoliometer measurements with the Cobb angles obtained by radiography, and to assess the sensitivity and specificity of the scoliometer measurements for the different diagnostic criteria for the referral of idiopathic scoliosis.METHODSixty-four patients were selected for the study: half with idiopathic scoliosis and half without. The 17 levels of the spine of each volunteer were measured with a scoliometer in the forward bending position. The measurements were performed three times on 42 volunteers by two different raters to obtain data for calculating the reliability values. Anteroposterior radiographs were taken to determine the Cobb angles, which were then compared with the highest trunk rotation value. Sensitivity and specificity were evaluated using radiograph criteria for referral: a Cobb angle of 10º and axial trunk rotation values between 5º and 10º.RESULTSExcellent intrarater reliability values and very good interrater reliability values were obtained. The correlation between the scoliometer measurements and radiograph analyses was considered good (r=0.7, p<0.05). The highest sensitivity value was for a trunk rotation of 5º at 87%.CONCLUSIONSThe scoliometer measurements showed a good correlation with the radiographic measurements.

BACKGROUND:Although photogrammetry is a widespread technique in the health field, despite of the methodological efforts distortions in the angular readings of the images are common.OBJECTIVE:To measure the error of angular measurements in photo images with different digital resolutions in an object with pre-determined angles.METHODS:We used a rubber ball with 52 cm in circumference. The object was previously marked with angles of 10°, 30°, 60° and 90° degrees. The photographic records were performed with the focal axis of the camera perpendicular and three meters away from the object, without the use of optical zoom and a resolution of 3, 5 and 10 Megapixels (Mp). All photographic records were stored and a previously trained experimenter using the computer program ImageJ analyzed the angular values of each photo. The measurements were performed twice within a fifteen-days interval. Subsequently, we calculated the accuracy, relative error and error in degrees values, precision and the Intraclass Correlation Coefficient (ICC).RESULTS:When analyzing the angle of 10°, the average accuracy of measurements was higher for those records of 3 Mp resolution compared to 5 and 10 Mp resolutions. The ICC was considered excellent for all resolutions. With regards to the analyzed angles in photographic records, it was possible to verify that the 90-degree angle photographs were more accurate, had lower relative error and error in degrees, and were more precise, regardless of image resolution.CONCLUSION:The photographs records that were taken with a 3 Mp resolution provided great accuracy and precision measurements and lower errors values, suggesting to be the proper resolution to generate image of angles of 10º and 30º.

The purpose of the present study was to analyze the energy expenditure and heart rate (HR), expressed as mean and maximal heart rate (HRmax), along matches of an official soccer competition. Methods: Eighteen under-20 year old (U-20) soccer players from a first-division Brazilian soccer team were evaluated during 15 matches. The rela­tionship between HR and oxygen uptake (HR/VO2 relationship) was established, and a linear regression equation was developed for each individual player participating in the study. This equation was then used to determine the oxygen uptake and the correspond­ing energy expenditure based on the HR values measured during the games. The HR was recorded at 5-second intervals, which were clustered in phases of 15 minutes (0-15 min, 15-30 min e 30-45 min) for each half time (first half [FH] and second half [SH]). Results: The players’ oxygen uptake was 308.3 ± 11.9 LO2/game and the energy expenditure was 17.3±1.3 Kcal.min-1 and 1,542.9±125.1 Kcal/game. The results showed that there were significant differences between the HR and HRmax identified in the first half (FH 15-30 min) compared to the second half (SH 0-15 min and SH 30-45 min), and at FH 30-45 min compared to SH 0-15 min (p<0.05). Conclusions: We concluded that soccer should be considered as an extremely arduous activity for U-20 players due to the high energy expenditure during the matches of an official championship. We also concluded that, during soccer games with young players, FH 15-30 min could be considered the most intense phase, because the highest values of HR and HRmax usually occur at this phase.