In this work, we present a novel sampling-based path planning method, called\nSPRINT. The method finds solutions for high dimensional path planning problems\nquickly and robustly. Its efficiency comes from minimizing the number of\ncollision check samples. This reduction in sampling relies on heuristics that\npredict the likelihood that samples will be useful in the search process.\nSpecifically, heuristics (1) prioritize more promising search regions; (2) cull\nsamples from local minima regions; and (3) steer the search away from\npreviously observed collision states. Empirical evaluations show that our\nmethod finds shorter or comparable-length solution paths in significantly less\ntime than commonly used methods. We demonstrate that these performance gains\ncan be largely attributed to our approach to achieve sample efficiency.\n