Autonomous Driving with Deep Learning: A Survey of State-of-Art Technologies

Since DARPA Grand Challenges (rural) in 2004/05 and Urban Challenges in 2007, autonomous driving has been the most active field of AI applications. Almost at the same time, deep learning has made breakthrough by several pioneers, three of them (also called fathers of deep learning), Hinton, Bengio and LeCun, won ACM Turin Award in 2019. This is a survey of autonomous driving technologies with deep learning methods. We investigate the major fields of self-driving systems, such as perception, mapping and localization, prediction, planning and control, simulation, V2X and safety etc. Due to the limited space, we focus the analysis on several key areas, i.e. 2D and 3D object detection in perception, depth estimation from cameras, multiple sensor fusion on the data, feature and task level respectively, behavior modelling and prediction of vehicle driving and pedestrian trajectories.

Hat tip to @kargarisaac.

1 Like

Al based off a decades old artificial neural network arrangement that has never had its underlying assumptions critically examined.
If you have an artificial neural network of width 1 million, each neuron has 1 million weight parameters to connect back to the neurons in the prior layer.
And for those 1 million weights you get 1 bend in the response curve of the layer the neuron is in.
Good deal? I think the effectiveness of weight pruning would kind of suggest otherwise, or the so called lottery ticket hypothesis.

Perhaps a more principled approach to artificial neural networks could get you down to 1 or 2 weight parameters per bend in the response curve and then perhaps a sudden jump in the performance even using current hardware.