• Publications
  • Influence
Parallel Online Temporal Difference Learning for Motor Control
  • W. Caarls, E. Schuitema
  • Computer Science, Medicine
  • IEEE Transactions on Neural Networks and Learning…
  • 1 July 2016
This paper shows that TD learning can work effectively in real robotic systems as well, using parallel model learning and planning, and achieves a speedup of almost two orders of magnitude over regular TD control on simulated control benchmarks. Expand
Comparison of extremum seeking control algorithms for robotic applications
The purpose of this paper is to help engineers and researches to choose among the extremum seeking control techniques for robotic applications such as object grasping, active object recognition and viewpoint optimization by proposing the usage of the approximation based methods when the noise level is negligible. Expand
Learning while preventing mechanical failure due to random motions
In case of LEO, a bipedal walking robot that tries to optimize a walking motion, the MTBF can be increased by a factor of 108 compared to SARSA(λ), indicating that, in some cases, failures due to high frequency random motions can be prevented without decreasing the performance. Expand
Automated Design of Application-Specific Smart Camera Architectures
This work uses a programming model based on the concept of architecture independence through algorithm dependence to conduct an automated design space exploration of possible architectures, creating a Pareto front of optimal trade-offs between performance, area and power consumption. Expand
Distance metric approximation for state-space RRTs using supervised learning
This work uses the Iterative Linear Quadratic Regulator approach for estimating an approximation to the optimal cost and learns this cost using Locally Weighted Projection Regression and shows that the learnt function approximates the original cost with a reasonable tolerance and gives a tremendous speed up of a factor of 1000 over the actual computation time. Expand
Algorithmic skeletons for stream programming in embedded heterogeneous parallel image processing applications
This paper presents a C-like skeleton implementation language, PEPCI, that uses term rewriting and partial evaluation to specify skeletons for parallel C dialects, and provides a stream programming language that is better tailored to the user as well as the underlying architecture. Expand
Q-caching: an integrated reinforcement-learning approach for caching and routing in information-centric networks
This paper proposes a caching strategy, namely Q-caching, which leverages information that is already collected by the routing algorithm, which promotes content diversity in the network, reducing the load at custodians and average download times for clients. Expand
Benchmarking model-free and model-based optimal control
A comprehensive comparison of the performance of reinforcement learning and nonlinear model predictive control for an ideal system as well as for a system with parametric and structural uncertainties suggests that benefits can be obtained by combining these methods for real systems being subject to such uncertainties. Expand
Architecture Study for Smart Cameras
The range of smart cameras of the Philips laboratories and its software support for easy programming that were partly developed in close co-operation with the SmartCam project are focused on. Expand
Skeletons and Asynchronous RPC for Embedded Data- and Task Parallel Image Processing
This paper describes how to exploit task parallelism using an asynchronous remote procedure call (RPC) system, optimized for low-memory and sparsely connected systems such as smart cameras. Expand