Research

Safe and Intuitive Flying Car (SafeTX)

Sebastian Scherer — 2011-10-10T09:30:28-04:00

In this project we develop the autonomy and human interfaces to be able to operate a flying car safely without requiring special pilot skills. The key challenges are safe detection of landing sites and intuitive control of the vehicle.

Link to the homepage of this project.

Riverine Mapping

Sebastian Scherer — 2011-10-10T09:30:10-04:00

In this project we develop perception and motion planning algorithms for a micro aerial vehicle that explores a river to build a map of the environment. The key problems we address are GPS-denied localization and control, river detection, and obstacle avoidance.

Link to the homepage of this project.

Landing Site Search & Evaluation for Autonomous Medical Evacuation

Sebastian Scherer — 2011-10-10T09:29:35-04:00

In this project we developed an algorithm for autonomous medical evacuation. We used a full-size helicopter to collect sensor data and test our algorithm at several test sites around Pittsburgh. The algorithm has two stages: First we perform a coarse optimistic evaluation of the terrain. A second stage performs a fine evaluation using a 3D surface of the helicopter. The best landing site is then chosen based on a cost metric.
Link to the homepage of this project.

We demonstrated the first helicopter that can select its own landing sites.

Fast Obstacle Avoidance for UAVs

Sebastian Scherer — 2011-10-10T09:28:55-04:00

The goal of this project was to develop a helicopter that is able to fly between a series of spare waypoints autonomously. The altitude however is so low that there are many obstacles present. Some of the obstacles that we avoided were thin wires (6mm), trees, buildings and containers. The helicopter is completely autonomous. It senses the world with a laser rangefinder and a state estimator and sends velocity commands to a controller.

In this paper are more details about our approach.

Link to videos about collision avoidance
Link to some more videos

Obstacle Avoidance for Unmanned Rotorcraft

Sebastian Scherer — 2011-10-10T09:27:13-04:00

In this project we automated a small quad-rotor vehicle to perform interesting missions. In the first project we performed 3D obstacle avoidance in a small 300g payload. Since then we have started to work on adding more capabilities such as wire capture to perform missions of unlimited duration, and GPS-denied navigation close and inside buildings.
In this paper are more details about how we are able to quickly calculate the C-space and cost functions for planning.
Link to videos about collision avoidance

Wire-Capture

GPS-Denied Hovering

Neural Network Controller for Unmanned Aerial Vehicles

Sebastian Scherer — 2011-10-10T09:26:43-04:00

We developed an online adaptive earning hovering controller. It consists of a PD-controller with a neural network as the adaptive element. As the helicopter experiences unknown disturbances the controller will adapt and respond. We implemented this project on a Blade CX helicopter.

Learning Obstacle Avoidance Parameters

Sebastian Scherer — 2011-10-10T09:26:02-04:00

In this project we learned obstacle avoidance parameters for a reactive obstacle avoidance control law from human observation. A subject drove to a prescribed goal while trying to avoid obstacles. We then fit the parameters of a reactive algorithm to best match the training paths that avoids obstacles.

Paper about learning obstacle avoidance parameters
Link to the Homepage of the Project

Motion and Sensor Simulation for Tartan Racing(Urban Challenge)

Sebastian Scherer — 2011-10-10T09:25:17-04:00

The aim of this project was to develop a simulator for the Urban challenge. The simulator was used to test and develop the motion planning algorithms and aided in the data visualization. The simulator was structured in a distributed fashion to split processing among several computers. It has the ability to handle an arbitrary number of clients and performs collision and constraint checking.
Link to the Urban Challenge homepage.