A Review of Tracking, Prediction and Decision Making Methods for Autonomous Driving

Abstract

This literature review focuses on three important aspects of an autonomous car system: tracking (assessing the identity of the actors such as cars, pedestrians or obstacles in a sequence of observations), prediction (predicting the future motion of surrounding vehicles in order to navigate through various traffic scenarios) and decision making (analyzing the available actions of the ego car and their consequences to the entire driving context). For tracking and prediction, approaches based on (deep) neural networks and other, especially stochastic techniques, are reported. For decision making, deep reinforcement learning algorithms are presented, together with methods used to explore different alternative actions, such as Monte Carlo Tree Search.

36 pages, 11 figures

References in corpus (30)

• Continuous control with deep reinforcement learning
• Simple Online and Realtime Tracking
• Trust Region Policy Optimization
• Beyond Correlation Filters: Learning Continuous Convolution Operators for Visual Tracking
• Tracking without bells and whistles
• Multi-Modal Trajectory Prediction of Surrounding Vehicles with Maneuver based LSTMs
• How would surround vehicles move? A Unified Framework for Maneuver Classification and Motion Prediction
• IntentNet: Learning to Predict Intention from Raw Sensor Data
• Explaining How a Deep Neural Network Trained with End-to-End Learning Steers a Car
• TraPHic: Trajectory Prediction in Dense and Heterogeneous Traffic Using Weighted Interactions
• TNT: Target-driveN Trajectory Prediction
• Imagination-Augmented Agents for Deep Reinforcement Learning
• Performance Measures and a Data Set for Multi-Target, Multi-Camera Tracking
• The Predictron: End-To-End Learning and Planning
• EvolveGraph: Multi-Agent Trajectory Prediction with Dynamic Relational Reasoning
• Multi-Object Tracking with Multiple Cues and Switcher-Aware Classification
• Probabilistic Vehicle Trajectory Prediction over Occupancy Grid Map via Recurrent Neural Network
• Conditional Flow Variational Autoencoders for Structured Sequence Prediction
• Online Multi-Object Tracking Using CNN-based Single Object Tracker with Spatial-Temporal Attention Mechanism
• VectorNet: Encoding HD Maps and Agent Dynamics from Vectorized Representation
• Long-term Pedestrian Trajectory Prediction using Mutable Intention Filter and Warp LSTM
• Towards Multi-Object Detection and Tracking in Urban Scenario under Uncertainties
• Learning to Search with MCTSnets
• An LSTM Approach to Temporal 3D Object Detection in LiDAR Point Clouds
• Deep Tracking: Visual Tracking Using Deep Convolutional Networks
• Learning Non-Uniform Hypergraph for Multi-Object Tracking
• Online Multi-Object Tracking with Historical Appearance Matching and Scene Adaptive Detection Filtering
• Spatial-Temporal Block and LSTM Network for Pedestrian Trajectories Prediction
• Proximal Gradient Temporal Difference Learning: Stable Reinforcement Learning with Polynomial Sample Complexity
• Fast Dynamic Convolutional Neural Networks for Visual Tracking

Cited by in corpus (10)

• Milestones in Autonomous Driving and Intelligent Vehicles: Survey of Surveys
• TrAISformer -- A Transformer Network with Sparse Augmented Data Representation and Cross Entropy Loss for AIS-based Vessel Trajectory Prediction
• Multi-Modal Sensor Fusion and Object Tracking for Autonomous Racing
• Trajectory Prediction for Autonomous Driving: Progress, Limitations, and Future Directions
• Diverse Controllable Diffusion Policy with Signal Temporal Logic
• An Inverse Optimal Control Approach for Trajectory Prediction of Autonomous Race Cars
• Characterizing Structured versus Unstructured Environments based on Pedestrians' and Vehicles' Motion Trajectories
• Driving Style Recognition at First Impression for Online Trajectory Prediction
• A Socially Aware Reinforcement Learning Agent for The Single Track Road Problem
• Bayesian optimization approach for tracking the location and orientation of a moving target using far-field data