Benchmarking Multimodal Trajectory Prediction Algorithms for Autonomous Trucking Applications

Institut

Lehrstuhl für Fahrzeugtechnik

Typ

Bachelorarbeit / Semesterarbeit /

Inhalt

experimentell / theoretisch /  

Beschreibung

Autonomous trucking depends on an accurate representation and understanding of the environment. This perception task does not only consist of object detection and localization but also tracking and prediction. However, to this date, the transferability of prediction algorithms for autonomous passenger cars to the autonomous trucking domain has not yet been shown. Therefore, state-of-the-art prediction algorithms should be implemented and benchmarked in autonomous trucking applications.

This Thesis offers a unique opportunity to implement state-of-the-art prediction algorithms and evaluate them on newly released autonomous trucking data for the first time. It aims to establish a trajectory prediction benchmark for autonomous trucking applications and evaluate the transferability of prediction methods for autonomous passenger cars. The goal is to develop an objective, reliable, and valid multimodal trajectory prediction benchmark for autonomous trucking and test it with state-of-the-art algorithms on real-world trucking data.

The first step of this project consists of literature research on the current state of the art in trajectory prediction and benchmarking. Afterwards, a prediction benchmark should be designed that meets the quality criteria of empirical research before different prediction algorithms should be implemented and tested in a third step. Finally, the results of the work should be compared to established benchmarks to derive future directives.

Work packages

• Literature research on trajectory prediction and benchmarking
• Design of a multimodal trajectory prediction benchmark
• Implementation of state-of-the-art prediction algorithms
• Evaluation of their suitability for autonomous trucking applications
• Comparison of the results and deduction of future directives

Voraussetzungen

Requirements

• Enthusiastic about autonomous driving
• Excited about Python programming
• Involved working attitude

 

The thesis can be written in German or English language.

 

Tags

FTM Studienarbeit, FTM AV, FTM AV Perception, FTM Informatik, FTM Fent

Möglicher Beginn

sofort

Kontakt

Felix Fent, M. Sc.
Raum: MW3508
Tel.: +49 89 289 15347
fentftm.mw.tum.de

Ausschreibung