Drift Reduced Navigation with Deep Explainable Features

Modern autonomous vehicles (AVs) often rely on vision, LIDAR, and even radar-based simultaneous localization and mapping (SLAM) frameworks for precise localization and navigation. However, modern SLAM frameworks often lead to unacceptably high levels of drift (i. e. , localization error) when AVs observe few visually distinct features or encounter occlusions due to dynamic obstacles. This paper argues that minimizing drift must be a key desiderata in AV motion planning, which requires an AV to take active control decisions to move towards feature-rich regions while also minimizing conventional control cost. To do so, we first introduce a novel data-driven perception module that observes LIDAR point clouds and estimates which features/regions an AV must navigate towards for drift minimization. Then, we introduce an interpretable model predictive controller (MPC) that moves an AV toward such feature-rich regions while avoiding visual occlusions and gracefully trading off drift and control cost. Our experiments on challenging, dynamic scenarios in the state-of-the-art CARLA simulator indicate our method reduces drift up to 76. 76% compared to benchmark approaches.

Authors

• Mohammad Omama
• Sundar Sripada V. S.
• Sandeep Chinchali
• Arun Kumar Singh 0001
• K. Madhava Krishna

Keywords

• Location awareness
• Visualization
• Simultaneous localization and mapping
• Laser radar
• Costs
• Navigation
• Heuristic algorithms
• Point Cloud
• Autonomous Vehicles
• Path Planning
• Model Predictive Control
• Cost Control
• LiDAR Point Clouds
• Perception Module
• Loss Function
• Positive Samples
• Negative Samples
• Latent Space
• Lateral Position
• Semantic Features
• Trajectory Optimization
• Semantic Differential
• Run Length
• Direct Losses
• Homotopy
• 3D Point Cloud
• Triplet Loss
• Proximal Policy Optimization
• Edge Features
• Dynamic Scenes
• Side Of The Road
• Planning Module
• Model Predictive Control Algorithm
• Vertical Field Of View
• Self-driving
• Convolutional Layers