The Experts below are selected from a list of 51 Experts worldwide ranked by ideXlab platform
Angela P Schoellig - One of the best experts on this subject based on the ideXlab platform.
-
visual localization with google earth images for robust global pose estimation of uavs
International Conference on Robotics and Automation, 2020Co-Authors: Bhavit Patel, Timothy D Barfoot, Angela P SchoelligAbstract:We estimate the global pose of a multirotor UAV by visually localizing images captured during a flight with Google Earth images pre-rendered from known poses. We metrically localize real images with georeferenced rendered images using a dense mutual Information Technique to allow accurate global pose estimation in outdoor GPS-denied environments. We show the ability to consistently localize throughout a sunny summer day despite major lighting changes while demonstrating that a typical feature-based localizer struggles under the same conditions. Successful image registrations are used as measurements in a filtering framework to apply corrections to the pose estimated by a gimballed visual odometry pipeline. We achieve less than 1 m and 1◦ RMSE on a 303 m flight and less than 3 m and 3◦ RMSE on six 1132 m flights as low as 36 m above ground level conducted at different times of the day from sunrise to sunset.
Farshid Keynia - One of the best experts on this subject based on the ideXlab platform.
-
day ahead price forecasting of electricity markets by mutual Information Technique and cascaded neuro evolutionary algorithm
IEEE Transactions on Power Systems, 2009Co-Authors: Nima Amjady, Farshid KeyniaAbstract:In a competitive electricity market, price forecasts are important for market participants. However, electricity price is a complex signal due to its nonlinearity, nonstationarity, and time variant behavior. In spite of much research in this area, more accurate and robust price forecast methods are still required. In this paper, a combination of a feature selection Technique and cascaded neuro-evolutionary algorithm (CNEA) is proposed for this purpose. The feature selection method is an improved version of the mutual Information (MI) Technique. The CNEA is composed of cascaded forecasters where each forecaster consists of a neural network (NN) and an evolutionary algorithm (EA). An iterative search procedure is also incorporated in our solution strategy to fine-tune the adjustable parameters of both the MI Technique and CNEA. The price forecast accuracy of the proposed method is evaluated by means of real data from the Pennsylvania-New Jersey-Maryland (PJM) and Spanish electricity markets. The method is also compared with some of the most recent price forecast Techniques.
Ross Berbeco - One of the best experts on this subject based on the ideXlab platform.
-
su ee a3 05 mutual Information for beams eye view lung tumor tracking without radiopaque markers
Medical Physics, 2009Co-Authors: Joerg Rottmann, M Aristophanous, So Yeon Park, Ross BerbecoAbstract:Purpose: To keep safety margins in lung stereotactic body radiation therapy(SBRT) small and provide retrospective calculation of the delivereddose, the tumor motion should be monitored. We propose a tumor tracking algorithm that can estimate the tumor location from portal images taken during the treatment without the help of fiducial markers. Method and Materials: An algorithm based on a normalized mutual Information Technique was developed for tumor tracking. First a tumor template and a search region are identified on a DRR set reconstructed from a 4DCT acquired prior to the treatment. The set consists of 10 images relating to 10 equally sized breathing phase bins. The template is then used to track the tumor over the sequence of portal images. To estimate the tracking precision a dynamic thorax phantom was employed. Results: The phantom study showed a sub millimeter tracking accuracy in the superior‐ inferior direction for anterior‐posterior and lateral fields. In a preliminary retrospective patient study the algorithm was able to track the tumor motion throughout the whole imagesequence. Manual verification yielded a tracking magnitude error of xy = (3.4 ± 0.8) mm. Furthermore the algorithm's robustness was tested with portal imagesequences from two other patients with different tumor motion amplitude and contrast. The accuaracy was estimated by comparison with manual tracking and yielded xy = (1.7 ± 1.9) mm and xy = (1.2 ± 0.8) mm, respectively. Conclusion: The algorithm has shown great potential for markerless lungtumor tracking. First test results showed that it can perform tumor tracking on portal images and DRRs even if the tracking template was defined in the other modality respectively. Conflict of Interest: Varian Medical Systems, Inc.
Bhavit Patel - One of the best experts on this subject based on the ideXlab platform.
-
visual localization with google earth images for robust global pose estimation of uavs
International Conference on Robotics and Automation, 2020Co-Authors: Bhavit Patel, Timothy D Barfoot, Angela P SchoelligAbstract:We estimate the global pose of a multirotor UAV by visually localizing images captured during a flight with Google Earth images pre-rendered from known poses. We metrically localize real images with georeferenced rendered images using a dense mutual Information Technique to allow accurate global pose estimation in outdoor GPS-denied environments. We show the ability to consistently localize throughout a sunny summer day despite major lighting changes while demonstrating that a typical feature-based localizer struggles under the same conditions. Successful image registrations are used as measurements in a filtering framework to apply corrections to the pose estimated by a gimballed visual odometry pipeline. We achieve less than 1 m and 1◦ RMSE on a 303 m flight and less than 3 m and 3◦ RMSE on six 1132 m flights as low as 36 m above ground level conducted at different times of the day from sunrise to sunset.
Nima Amjady - One of the best experts on this subject based on the ideXlab platform.
-
day ahead price forecasting of electricity markets by mutual Information Technique and cascaded neuro evolutionary algorithm
IEEE Transactions on Power Systems, 2009Co-Authors: Nima Amjady, Farshid KeyniaAbstract:In a competitive electricity market, price forecasts are important for market participants. However, electricity price is a complex signal due to its nonlinearity, nonstationarity, and time variant behavior. In spite of much research in this area, more accurate and robust price forecast methods are still required. In this paper, a combination of a feature selection Technique and cascaded neuro-evolutionary algorithm (CNEA) is proposed for this purpose. The feature selection method is an improved version of the mutual Information (MI) Technique. The CNEA is composed of cascaded forecasters where each forecaster consists of a neural network (NN) and an evolutionary algorithm (EA). An iterative search procedure is also incorporated in our solution strategy to fine-tune the adjustable parameters of both the MI Technique and CNEA. The price forecast accuracy of the proposed method is evaluated by means of real data from the Pennsylvania-New Jersey-Maryland (PJM) and Spanish electricity markets. The method is also compared with some of the most recent price forecast Techniques.