Deep learning has been used to demonstrate end-to-end neural network learningfor autonomous vehicle control from raw sensory input . We present an efficient and robust LiDAR-based end- to-end navigation framework . We evaluate our system on a full-scale vehicle anddemonstrate lane-stable as well as navigation capabilities . In the presence ofout-of-distribution events (e.g., sensor failures), our system significantlyimproves robustness and reduces the number of takeovers in the real world. We then propose Hybrid Evidential Fusion that directly estimates the uncertainty of theprediction from only a single forward pass and then fuses the controlpredictions intelligently. We evaluated our system to a full scale vehicle and demonstrated lane-stability and navigation capabilities.
Author(s) : Zhijian Liu, Alexander Amini, Sibo Zhu, Sertac Karaman, Song Han, Daniela RusLinks : PDF - Abstract
Code :
https://github.com/oktantod/RoboND-DeepLearning-Project
Keywords : navigation - vehicle - system - lidar - robust -
- Kaggle Competitions
- Google Cloud Professional Data Engineer Specialization
- School of Data Science
- Introduction to TensorFlow for AI, DL and ML by Andrew Ng
- How to Win a Data Science Competition on Coursera
- Mathematics for Machine Learning by Marc Peter Deisenroth, A. Aldo Faisal and Cheng Soon Ong
- fast.ai Machine Learning
- Mathematics for Machine Learning Specialization