Bertil Svensson

The VEGA moderately parallel MIMD, moderately parallel SIMD, architecture for high performance array signal processing

In array radar signal processing applications, the processing demands range from tens of GFLOPS to several TFLOPS. To address this, as well as the, size and power dissipation issues, a special purpose array signal processing architecture is proposed. We argue that a combined MIMD-SIMD system can give flexibility, scalability, and programmability as well as high computing density. The MIMD system level, where SIMD modules are interconnected by a fiber-optic real-time network, provides the high level flexibility while the SIMD module level provides the compute density. In this paper we evaluate different design alternatives and show how the VEGA architecture was derived. By examining the applications and the algorithms used, the SIMD mesh processor is found be sufficient. However, the smaller the meshes are the better is the flexibility and efficiency. Then, based on prototype VLSI implementations and on instruction statistics, we find that a relatively large pipelined processing element maximises the performance per area. It is thereby concluded that the small SIMD mesh processor array with powerful processing elements is the best choice. These observations are further exploited in the design of the single-chip SIMD processor array to be included in the MIMD-style overall system. The system scales from 6.4 GFLOPS to several TFLOPS peak performance.

Neural Virtual Sensors — Estimation of Combustion Quality in SI Engines using the Spark Plug

We propose two virtual sensors which estimate the location of the pressure peak and the air-fuel ratio from measurements of the ionization current across the spark plug gap.

Self-orienting with on-line learning of environmental features

Evidence from recently conducted neurophysiological experiments on freely moving rats has revealed that the firing of the head-direction cell ensemble predicts the future head direction in response to the vestibular input and that visual cues strongly influence the shift of the tuning curve represented by the firing of the head-direction cell ensemble. In this article, we investigate the possibility of using learned landmark features to self-orient an autonomous agent in a partially known environment. A model is suggested that incorporates an artificial head-direction system for emulating the behavior of head-direction cell ensembles in biological systems, a lattice-based dynamic cell structure for categorizing and classifying environmental features, and an expectancy-based learning mechanism that learns to associate each head direction with a certain environmental feature. Our experimental results show that the suggested model is capable of correcting the drift in the orientation estimated by dead-reckoning.

A Low-Risk Approach to Mobile Robot Path Planning

This paper presents a self-organizing approach for mobile robot path planning problems in dynamic environments by using case-based reasoning together with a more conventional method of grid-map based path planning. The map-based path planner is used to suggest new innovative solutions for a particular path planning problem. The case-base is used to store the paths and evaluate their traversability. While planning the route those paths are preferred which, according to former experience, are least risky. As the environment changes, the exploration as well as the evaluation of the paths will allow the system to self-organize by forming a set of low-risk paths that are safest to follow. The experiments in a simulated environment show that the robot is able to adapt in a dynamic environment and learns to use the least risky paths.

A processor array module for distributed, massively parallel, embedded computing

With the increased degree of miniaturization resulting from the use of modem VLSI technology and the high communication bandwidth available through optical connections, it is now possible to build ...

Using Focus to Direct Environmental Mapping

Biological data reveal that the activation of hippocampal place cells is highly correlated to the process of landmark detection when animals are performing navigational tasks. As a functional approximation to hippocampal place learning, this paper presents a network model that can be used by an autonomous agent to map landmarks, places and the spatial relation between them. For the network to be used to map a large space, a focusing mechanism is introduced in the network to direct environmental mapping and to limit the amount of computation needed.

Effective mapping of artificial neural network algorithms onto massively parallel hardware: the REMAP programming environment

The application of artificial neural networks (ANN) in real-time embedded systems demands high performance computers. Miniaturized massively parallel architectures are suitable computation platforms for this task. An important question which arises is how to establish an effective mapping from ANN algorithms to hardware. In this paper, we demonstrate how an effective mapping can be achieved with our programming environment in close combination with an optimized architecture design targeted for neuro-computing.<<ETX>>