Robohub.org
 

Learning Bayesian filters without precise ground truth

by
21 January 2011



share this:

As seen in a previous post, Bayes filters such as Kalman filters can be used to estimate the state of a robot. Usually, this requires having a model of how the robot’s sensor measurements relate to the state you want to observe and a model to predict how the robot’s control inputs impact its state. When little is know about these models, machine learning techniques can help find models automatically.

In the scenario imagined by Ko et al. the goal is to infer a car’s position (state) on a race track based on remote control commands and measurements from an Inertial Measurement Unit (IMU, see red box in picture below) that provides turn rates in roll, pitch, and yaw and accelerations in 3 dimensions. The car is only controlled in speed since a “rail” in the track makes it turn. As a start, Ko et al. collect data by driving the car around the track while recording its remote control commands, IMU measurements, and position (ground truth) estimated using an overhead camera. The data is then used to train probabilistic models (Gaussian Process models) that are finally integrated into a Bayes filter.

However, the need for ground truth requires extra effort and additional hardware such as the overhead camera. To overcome this problem, Ko et al. extend their method to deal with situations where no, little, or noisy ground truth is used to train the models.

Results show the successful tracking of the car with better performance than state-of-the-art approaches, even when no ground truth is given. The authors also show how the developed method can be used to allow the car or a robot arm to replay trajectories based on expert demonstrations.




Sabine Hauert is President of Robohub and Associate Professor at the Bristol Robotics Laboratory
Sabine Hauert is President of Robohub and Associate Professor at the Bristol Robotics Laboratory





Related posts :



Robot Talk Episode 42 – Thom Kirwan-Evans

In this week's episode of the Robot Talk podcast, host Claire Asher chatted to Thom Kirwan-Evans from Origami Labs all about computer vision, machine learning, and robots in industry.
25 March 2023, by

Resilient bug-sized robots keep flying even after wing damage

New repair techniques enable microscale robots to recover flight performance after suffering severe damage to the artificial muscles that power their wings.
23 March 2023, by

How drones for organ transportation are changing the healthcare industry

The healthcare drone industry has witnessed a dramatic surge in the last couple of years. In 2020, the market grew 30% and is expected to grow from $254 million in 2021 to $1,5 billion in 2028.
21 March 2023, by

Robotic bees and roots offer hope of healthier environment and sufficient food

Miniature robots that mimic living organisms are being developed to explore and support real-life ecosystems.
18 March 2023, by

Robot Talk Episode 41 – Alessandra Rossi

In this week's episode of the Robot Talk podcast, host Claire Asher chatted to Alessandra Rossi from the University of Naples all about social robotics, theory of mind, and robots playing football.
17 March 2023, by





©2021 - ROBOTS Association


 












©2021 - ROBOTS Association