Robots can collide into obstacles they have imperfect information about. We aim to allieviate this by introducing a model of uncertainty in obstacle geometry and presenting a "safe" planner. For any specified \(\epsilon\), the output of the safe planner collides with probability at most \(\epsilon\). The planner can extend these guarantees to dynamical systems using sum of squares funnels.Coming Soon!
We develop a new paradigm for designing fully streaming, area-efficient FPGA implementations of common building blocks for vision algorithm. By focusing on avoiding redundant computation we achieve a reduction of one to two orders of magnitude reduction in design area utilization as compared to previous implementations. We demonstrate that our design works in practice by building five 325 frames per second, high resolution Harris corner detection cores onto a single FPGA.View PDF