NMR: Neural Manifold Representation for Autonomous Driving
This work addresses a specific limitation in autonomous driving systems for improved path planning, but it appears incremental as it builds on existing BEV representations by adapting them to handle road gradients.
The paper tackles the problem of distorted Bird's-Eye View (BEV) representations in autonomous driving caused by non-coplanar road gradients, which leads to inefficient and incorrect path planning. It proposes Neural Manifold Representation (NMR), a method that learns to infer semantics and predict way-points on a manifold, resulting in motion plans consistent with surface geometry, with plans to test on CARLA and SYNTHIA-SF datasets.
Autonomous driving requires efficient reasoning about the Spatio-temporal nature of the semantics of the scene. Recent approaches have successfully amalgamated the traditional modular architecture of an autonomous driving stack comprising perception, prediction, and planning in an end-to-end trainable system. Such a system calls for a shared latent space embedding with interpretable intermediate trainable projected representation. One such successfully deployed representation is the Bird's-Eye View(BEV) representation of the scene in ego-frame. However, a fundamental assumption for an undistorted BEV is the local coplanarity of the world around the ego-vehicle. This assumption is highly restrictive, as roads, in general, do have gradients. The resulting distortions make path planning inefficient and incorrect. To overcome this limitation, we propose Neural Manifold Representation (NMR), a representation for the task of autonomous driving that learns to infer semantics and predict way-points on a manifold over a finite horizon, centered on the ego-vehicle. We do this using an iterative attention mechanism applied on a latent high dimensional embedding of surround monocular images and partial ego-vehicle state. This representation helps generate motion and behavior plans consistent with and cognizant of the surface geometry. We propose a sampling algorithm based on edge-adaptive coverage loss of BEV occupancy grid and associated guidance flow field to generate the surface manifold while incurring minimal computational overhead. We aim to test the efficacy of our approach on CARLA and SYNTHIA-SF.