Welcome to the ALR-Lab                                       

The Autonomous Learning Robots (ALR) Lab at the Institute for Anthropomatics and Robotics External Link of the Department of Informatics External Link, focuses on the development of novel machine learning methods for robotics.  Future robot technology will have to deal with very challenging real world scenarios that are quite different from the lab environments typically considered in robotics research. Real world environments are unknown and unstructured, consisting of objects of unpredictable shapes or even other, unknown agents such as humans. The robot can encounter so many different situations while interacting with such environments that pre-programming such tasks seems to be infeasible.

  

Our research is focused on the intersection of machine learning, robotics, human-robot interaction and computer vision. Our goal is to create data-efficient and mathematically principled machine learning algorithms that are suitable for complex robot domains such as grasping and manipulation, forceful interactions or dynamic motor tasks. In our research, we always aim for a strong theoretical basis for our developed algorithms which are derived from first order principles. In terms of methods, our work is focused on:

  • Movement Representations
  • Reinforcement Learning and Policy Search
  • Imitation Learning and Interactive Learning
  • Model-Learning
  • Perception

While we thrive to extend the state of the art for each of these areas of machine learning, our vision is to create an orchestration of these methods in order to develop a fully autonomous learning robotics system. 

News:

CoRL 22: Inferring Versatile Behavior from Demonstrations by Matching Geometric Descriptors

We combine a mixture of movement primitives with a distribution matching objective to learn versatile behaviors that match the expert’s behavior and versatility.

more
New TMLR paper: On Uncertainty in Deep State Space Models for Model-Based Reinforcement Learning

We study how recent State Space modeling approaches for Model-Based RL represent uncertainties. We find some flaws and propose a theoretically better-grounded alternative. We show it improves performance in tasks where it is important to appropriately capture uncertainty. If you want to know who this relates to the cat and the hamster you have to read the paper.

more
CoRL 22: Deep Black-Box Reinforcement Learning with Movement Primitives

Episode-based reinforcement learning (ERL) algorithms treat reinforcement learning (RL) as a black-box optimization problem where we learn to select a parameter vector of a controller, often represented as a movement primitive, for a given task descriptor called a context.

more
New Paper at ECCV 2022 about Deep Hierarchical Variational Autoencoding for RGB Image Fusion

We present a novel deep hierarchical variational autoencoder that can serve as a basis for many fusion tasks. It can generate diverse image samples that are conditioned on multiple noisy, occluded, or only partially visible input images. We created three novel image fusion datasets and show that our method outperforms traditional approaches significantly.

more
New Paper at IROS 2022 about Multi-View 6D Pose Estimation on RGB-D Frames

We present a novel deep learning method that estimates the 6D poses of all objects in a cluttered scene based on multiple RGB-D images. Our approach is considerably more accurate than previous approaches especially for very occluded objects and it is robust towards dynamic camera setups as well as inaccurate camera calibration.

more
.
New RSS 2022 Paper: End-to-End Learning of Hybrid Inverse Dynamics Models

We propose a new formulation for a residual hybrid inverse dynamics model, which combines a fully physically consistent rigid-body dynamics model with a recurrent LSTM and a Coulomb friction function. The model is trained end-to-end using a new formulation of Barycentric Parameters called “Differentiable Barycentric”, which implicitly guarantees all conditions of physical consistency. In our real robot motion tracking experiments we show, that the new model is able to achieve compliant and precise motion tracking on unseen movements.

more
.
New ICRA Paper: Push-2-See

In this paper, we present a new approach for interactive scene segmentation using deep reinforcement learning. Our robot can learn to push objects in a heap such that semantic segmentation algorithms can detect every object in the heavily cluttered heap. 

 

more
.
New ICRA Paper: Hierarchical Policy Learning for Mechanical Search

Mechanical Search (MS)  is a framework for object retrieval, which uses a heuristic algorithm for pushing and rule-based algorithms for high-level planning. While rule-based policies profit from human intuition in how they work, they usually perform sub-optimally in many cases. We present am deep hierarchical reinforcement learning (RL) algorithm to perform this task, showing an increased search performance in terms of number of needed manipulation, success rate as well as computation time!

more
New CVPR 2022 paper: What Matters for Meta-Learning Vision Regression Tasks?

We design two new types of cross-category level vision regression tasks, namely object discovery and pose estimation, which are of unprecedented complexity in the meta-learning domain for computer vision with exhaustively evaluation of common meta-learning techniques to strengthen the generalization capability. Furthermore, we propose functional contrastive learning (FCL) over the task representations in Conditional Neural Processes (CNPs) and train in an end-to-end fashion.

more
New ICLR Paper Accepted: Hidden Parameter State Space Models

We propose a multi-task deep Kalman model, that can adapt to changing dynamics and environments. The model gives state of the art performance on several robotic benchmarks with non-stationarity with little computational overhead!!!

more
versatile skills
New Paper @ CoRL 21: Specializing Versatile Skill Libraries Using Local Mixture of Experts

We propose a new method which enables robots to learn versatile and highly accurate skills in the contextual policy search setting by optimizing a mixture of experts model. We make use of Curriculum Learning, where the agent concentrates on local context regions it favors. Mathematical properties allow the algorithm to adjust the model complexity to find as many solutions as possible.

A video presenting our work can be found here.

more
New Paper @ IROS: Residual Feedback Learning for Contact-Rich Manipulation Tasks with Uncertainty
New Paper @ IROS 21: Residual Feedback Learning for Contact-Rich Manipulation Tasks with Uncertainty

We developped a new residual reinforcement learning method that not just manipulated the output of a controller but also its input (e.g., the set-points). We applied this method to a real robot peg-in-the hole setup with a significant amount of position and orientation uncertainty.

Video

more
New ICLR paper! Differentiable Trust Region Layers for Deep Reinforcement Learning
New ICLR paper! Differentiable Trust Region Layers for Deep Reinforcement Learning

Do you like Deep RL methods such as TRPO or PPO? Then you will also like this one! Our differentiable trust region layers can be used on top of any policy optimization algorithms such as policy gradients to obtain stable updates -- no approximations or implementation choices required :) Performance is enpar with PPO on simple exploration scenarios while we outperform PPO on more complex exploration environments.

New ICLR paper! Bayesian Context Aggregation for Neural Processes
New ICLR paper! Bayesian Context Aggregation for Neural Processes

Neural Processes are powerful tools for probabilistic meta-learning. Yet, they use rather basic aggregation methods, i.e. using a mean aggregator for the context, which does not give consistent uncertainty estimates and leads to poor prediction performance. Aggregating in a Bayesian way using Gaussian conditioning does a much better job !:)

CoRL paper accepted - AC-RKN For Dynamics Learning
CoRL 2021: paper accepted - Action Conditional Recurrent Kalman Networks

Action conditional probabilistic model inspired by Kalman filter operations in the latent state. Find out how we learn the complex non-markovian dynamics of pneumatic soft robots and large hydraulic robots with this disentangled state and action representation.

more
.
New JMLR Paper - Need to approximate complex distributions with a GMM? Here you go!

Many methods for machine learning rely on approximate inference from intractable probability distributions. Learning sufficiently accurate approximations requires a rich model family and careful exploration of the relevant modes of the target distribution...

more
Expected Information Maximization
ICLR 2020: paper accepted - Expected Information Maximization

 Using the I-Projection for Mixture Density Estimation. Find out why maximum likelihood is not well suited for mixture density modelling and why you should use the I-projection instead.  

more
ALR Logo
Starting @ KIT

The Autonomous Learning Robots (ALR) Lab was founded at Jan. 2020 at the KIT. The new group is now building up and looking forward to do exciting research and teaching!