J.J. Brickley
Papers
1
Total Citations
2
H-Index
1
About
J.J. Brickley’s research lies at the intersection of autonomous robotics, sensor fusion, and neural network clustering, with a focus on enabling mobile robots to perceive and navigate uncertain environments. Their most cited work introduces a novel integration of a hyper-ellipsoid clustering (HEC) Kohonen neural network with the Julier-Uhlmann-Kalman filter (JUKF) for simultaneous map building and tracking. This approach leverages the Mahalanobis distance to learn elongated, anisotropic shapes from sonar data—a critical advance for handling the noisy, non-spherical distributions typical of real-world sensor readings. By fusing self-organizing maps with probabilistic filtering, Brickley’s method provides a stochastic framework for low-level position estimation and environmental mapping, directly addressing the challenges of autonomous navigation. Though the work has garnered 2 citations, its conceptual contribution—bridging unsupervised clustering with Kalman-based state estimation—offers a foundational perspective for researchers exploring robust perception in robotics. Brickley’s research underscores the value of hybrid architectures that combine neural adaptability with statistical rigor, a theme that continues to resonate in modern autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1