Task-Relevant Failure Detection for Trajectory Predictors in Autonomous Vehicles

07/25/2022
by   Alec Farid, et al.
0

In modern autonomy stacks, prediction modules are paramount to planning motions in the presence of other mobile agents. However, failures in prediction modules can mislead the downstream planner into making unsafe decisions. Indeed, the high uncertainty inherent to the task of trajectory forecasting ensures that such mispredictions occur frequently. Motivated by the need to improve safety of autonomous vehicles without compromising on their performance, we develop a probabilistic run-time monitor that detects when a "harmful" prediction failure occurs, i.e., a task-relevant failure detector. We achieve this by propagating trajectory prediction errors to the planning cost to reason about their impact on the AV. Furthermore, our detector comes equipped with performance measures on the false-positive and the false-negative rate and allows for data-free calibration. In our experiments we compared our detector with various others and found that our detector has the highest area under the receiver operator characteristic curve.

READ FULL TEXT
research
07/02/2022

Long-Tail Prediction Uncertainty Aware Trajectory Planning for Self-driving Vehicles

A typical trajectory planner of autonomous driving usually relies on pre...
research
07/25/2019

Vehicular Multi-object Tracking with Persistent Detector Failures

Autonomous vehicles often perceive the environment by feeding sensor dat...
research
02/08/2017

Optimal Detection of Faulty Traffic Sensors Used in Route Planning

In a smart city, real-time traffic sensors may be deployed for various a...
research
02/11/2022

Failure Prediction with Statistical Guarantees for Vision-Based Robot Control

We are motivated by the problem of performing failure prediction for saf...
research
02/25/2020

Non-Gaussian Chance-Constrained Trajectory Planning for Autonomous Vehicles in the Presence of Uncertain Agents

Agent behavior is arguably the greatest source of uncertainty in traject...
research
08/24/2023

Interaction-Aware Trajectory Prediction and Planning in Dense Highway Traffic using Distributed Model Predictive Control

In this paper we treat optimal trajectory planning for an autonomous veh...
research
02/08/2022

Causal Scene BERT: Improving object detection by searching for challenging groups of data

Modern computer vision applications rely on learning-based perception mo...

Please sign up or login with your details

Forgot password? Click here to reset