Martijn Wisse

Pablo Lanillos, Cristian Meo, Corrado Pezzato et al.

Active inference is a mathematical framework which originated in computational neuroscience as a theory of how the brain implements action, perception and learning. Recently, it has been shown to be a promising approach to the problems of state-estimation and control under uncertainty, as well as a foundation for the construction of goal-driven behaviours in robotics and artificial agents in general. Here, we review the state-of-the-art theory and implementations of active inference for state-estimation, control, planning and learning; describing current achievements with a particular focus on robotics. We showcase relevant experiments that illustrate its potential in terms of adaptation, generalization and robustness. Furthermore, we connect this approach with other frameworks and discuss its expected benefits and challenges: a unified framework with functional biological plausibility using variational Bayesian inference.

Fred Bos, Ajith Anil Meera, Dennis Benders et al.

The free energy principle from neuroscience provides a brain-inspired perception scheme through a data-driven model learning algorithm called Dynamic Expectation Maximization (DEM). This paper aims at introducing an experimental design to provide the first experimental confirmation of the usefulness of DEM as a state and input estimator for real robots. Through a series of quadcopter flight experiments under unmodelled wind dynamics, we prove that DEM can leverage the information from colored noise for accurate state and input estimation through the use of generalized coordinates. We demonstrate the superior performance of DEM for state estimation under colored noise with respect to other benchmarks like State Augmentation, SMIKF and Kalman Filtering through its minimal estimation error. We demonstrate the similarities in the performance of DEM and Unknown Input Observer (UIO) for input estimation. The paper concludes by showing the influence of prior beliefs in shaping the accuracy-complexity trade-off during DEM's estimation.

Ajith Anil Meera, Martijn Wisse

The quest for a brain-inspired learning algorithm for robots has culminated in the free energy principle from neuroscience that models the brain's perception and action as an optimization over its free energy objectives. Based on this idea, we propose an estimation algorithm for accurate output prediction of a quadrotor flying under unmodelled wind conditions. The key idea behind this work is the handling of unmodelled wind dynamics and the model's non-linearity errors as coloured noise in the system, and leveraging it for accurate output predictions. This paper provides the first experimental validation for the usefulness of generalized coordinates for robot perception using Dynamic Expectation Maximization (DEM). Through real flight experiments, we show that the estimator outperforms classical estimators with the least error in output predictions. Based on the experimental results, we extend the DEM algorithm for model order selection for complete black box identification. With this paper, we provide the first experimental validation of DEM applied to robot learning.

Corrado Pezzato, Carlos Hernandez Corbato, Stefan Bonhof et al.

We propose a hybrid combination of active inference and behavior trees (BTs) for reactive action planning and execution in dynamic environments, showing how robotic tasks can be formulated as a free-energy minimization problem. The proposed approach allows handling partially observable initial states and improves the robustness of classical BTs against unexpected contingencies while at the same time reducing the number of nodes in a tree. In this work, we specify the nominal behavior offline, through BTs. However, in contrast to previous approaches, we introduce a new type of leaf node to specify the desired state to be achieved rather than an action to execute. The decision of which action to execute to reach the desired state is performed online through active inference. This results in continual online planning and hierarchical deliberation. By doing so, an agent can follow a predefined offline plan while still keeping the ability to locally adapt and take autonomous decisions at runtime, respecting safety constraints. We provide proof of convergence and robustness analysis, and we validate our method in two different mobile manipulators performing similar tasks, both in a simulated and real retail environment. The results showed improved runtime adaptability with a fraction of the hand-coded nodes compared to classical BTs.

Wouter Wolfslag, Mukunda Bharatheesha, Thomas Moerland et al.

Sampling-based kinodynamic planners, such as Rapidly-exploring Random Trees (RRTs), pose two fundamental challenges: computing a reliable (pseudo-)metric for the distance between two randomly sampled nodes, and computing a steering input to connect the nodes. The core of these challenges is a Two Point Boundary Value Problem, which is known to be NP-hard. Recently, the distance metric has been approximated using supervised learning, reducing computation time drastically. The previous work on such learning RRTs use direct optimal control to generate the data for supervised learning. This paper proposes to use indirect optimal control instead, because it provides two benefits: it reduces the computational effort to generate the data, and it provides a low dimensional parametrization of the action space. The latter allows us to learn both the distance metric and the steering input to connect two nodes. This eliminates the need for a local planner in learning RRTs. Experimental results on a pendulum swing up show 10-fold speed-up in both the offline data generation and the online planning time, leading to at least a 10-fold speed-up in the overall planning time.

Carlos Hernandez, Mukunda Bharatheesha, Wilson Ko et al.

This paper describes Team Delft's robot, which won the Amazon Picking Challenge 2016, including both the Picking and the Stowing competitions. The goal of the challenge is to automate pick and place operations in unstructured environments, specifically the shelves in an Amazon warehouse. Team Delft's robot is based on an industrial robot arm, 3D cameras and a customized gripper. The robot's software uses ROS to integrate off-the-shelf components and modules developed specifically for the competition, implementing Deep Learning and other AI techniques for object recognition and pose estimation, grasp planning and motion planning. This paper describes the main components in the system, and discusses its performance and results at the Amazon Picking Challenge 2016 finals.