Transportation

❯

Autonomous driving

❯

Autonomous apron truck

Automated transportation of luggage (carts) to requested destinations on an airport apron while following local traffic rules and resolving unplanned conflicts.

Goal:

Improve Operation Efficiency

Government

Defense & Military - Autonomous Vehicle

For:

Military Operatives.

Scope:

Using computer vision to implement perimeter safety for Autonomous Vehicles.

Goal:

Improved Product Development / R&D

AI components for vehicle platooning on public roads

Scope:

Trains of vehicles that drive very close to each other at nearly equal speed (platoons) on public roads, in particular platooning trucks on motorways.

Goal:

Improve Operation Efficiency

Autonomous trains (Unattended train operation (UTO))

Scope:

Freight and passenger trains operate autonomously, excluding any crew presence on board, but with remote operator attention involved.

Goal:

Improve Operation Efficiency

Self-driving aircraft towing vehicle

Scope:

Self-Driving towing vehicle for aircrafts, operating on an airfield autonomously.

Goal:

Improve Operation Efficiency

Autonomous apron truck

For:

Transporter

Goal:

Improve Operation Efficiency

Problem addressed

Automate transport to increase reliability, precision, efficiency and safety.

Scope of use case

Automated transportation of luggage (carts) to requested destinations on an airport apron while following local traffic rules and resolving unplanned

conflicts.

Description

While the number of airplanes visiting German airports has

steadily increased over the last decades and recently reached

a new all-time high the logistics to enable smooth processing

has also increased correspondingly in complexity.

To further manage an even higher number of airplanes a fully

automated luggage truck has been developed.

The truck receives tasks from a machine or human

coordinator and automatically execute these. For specific

tasks, such as loading and unloading or maintenance, further

interaction with human workers is needed. Therefore, the

truck is able to communicate its status and intents to

surrounding workers.

While operating on the apron the truck always obeys local

traffic rules. The only occasion on which these rules are

violated is if an accident is thereby avoided. Human safety is

always the trucks first priority.

For achieving all these functions an AI system consisting of

multiple individual elements, which are all expected to

operate collaboratively has been designed. The three main

modules are a perception module, a behaviour generator and

an execution module.

The truck perceives its environment by its perception

module, which consists of multiple submodules, such as

object detection, recognition, tracking and data fusion blocks

for multiple sensor types. The perceived information and the

respective uncertainties are further processed to localize, re-

project and detect each objects intent in the trucks

coordinate system.

The perception unit outputs a context model that the

behaviour generator receives to decide on what actions to

take next. This behaviour generator consists of a deep

reinforcement learning agent and is supervised by a

symbolic rule checker to ensure that the agent operates fault

free. If a taken action violates a rule, either the agent is

expected to determine a new action or, in safety critical

situations, the rule checker determines a safe action by

symbolic reasoning.

The execution module executes the behaviour determined by

the behaviour generator. It consists of motion planning,

control and communication submodules that execute the

intended task while reporting back to the behaviour

generator, which react to unexpected situations.

Additionally, the trucks status and intents are constantly

reported over communication systems to its surroundings to

enable uncomplicated interaction with the truck.

Live Video

Computer Vision

Interested in the same or similar project?
Submit a request and get a free project evaluation.