Jean Baillie · 2005
The growing diversity and complexity of existing robotic devices like humanoids, animal-like robots or wheeled robots, lead to the development of several incompatible software interfaces to control these robots. We believe that there is a need for a standard which could be universal, powerful and easy to use. The open source project URBI, a universal robotic body interface, aims at providing the ground for such a standard. It is based on a client/server architecture where the server is running on the robot and accessed by the client, on the robot or remotely via TCP/IP. The URBI language is a scripted language used by the client and capable of controlling the joints of the robot or access its sensors, camera, speakers or any accessible part of the machine. We present in this article an introduction to URBI and we describe how URBI differs from currently existing solutions. As an example making use of URBI-specific features, we present a simple perturbative approach to walk pattern generation, with URBI running on a ERS7 Aibo robot.
Cited by 109
Amitabh Sinha, Reyaansh Sinha · Zenodo (CERN European Organization for Nuclear Research) · 2026
This repository contains the research paper for Splice, an experimental programming language designed for embedded and resource-constrained systems. The project explores how a lightweight programming language and virtual machine architecture can provide a balance between simplicity, portability, and runtime efficiency. Splice was designed to offer a more accessible alternative to traditional embedded programming languages while maintaining practical performance. The paper discusses the motivation behind the language, its overall architecture, and benchmark comparisons with several existing scripting environments used in lightweight computing systems. This work represents an early-stage exploration of programming language design for embedded environments and aims to contribute ideas toward simpler and more efficient development tools for small devices.
Cited by 0Open access
Amitabh Sinha, Reyaansh Sinha · Zenodo (CERN European Organization for Nuclear Research) · 2026
This repository contains the research paper for Splice, an experimental programming language designed for embedded and resource-constrained systems. The project explores how a lightweight programming language and virtual machine architecture can provide a balance between simplicity, portability, and runtime efficiency. Splice was designed to offer a more accessible alternative to traditional embedded programming languages while maintaining practical performance. The paper discusses the motivation behind the language, its overall architecture, and benchmark comparisons with several existing scripting environments used in lightweight computing systems. This work represents an early-stage exploration of programming language design for embedded environments and aims to contribute ideas toward simpler and more efficient development tools for small devices.
Cited by 0Open access
Chia-Chu Chiang · 2004
Low-level language constructs used for expressing explicit communication, concurrency, synchronization, and parallelism in distributed systems for parallel processing are considered harmful. We are providing a programming model to support implicit communication, concurrency, synchronization, and parallelism in systems through an implicit coordination-oriented approach. A 4-layered interconnection architecture will be implemented to facilitate programming in an integrated manner. The implicit coordination-oriented approach to supporting parallel programming provides a number of benefits. Without inserting the low-level language constructs in an unstructured manner in programs makes the programs modular. Modularity improves the maintainability of the programs. Our approach supports the portability of programs by allowing the programs in different programming languages to be executed in any general programming environment without modifications.
Cited by 3
Yaoyao Ding, Bohan Hou, Xiao Zhang et al. · ArXiv.org · 2025
Serving Large Language Models (LLMs) is critical for AI-powered applications, yet it demands substantial computational resources, particularly in memory bandwidth and computational throughput. Low-precision computation has emerged as a key technique to improve efficiency while reducing resource consumption. Existing approaches for generating low-precision kernels are limited to weight bit widths that are powers of two and suffer from suboptimal performance because of high-level GPU programming abstractions. These abstractions restrict critical optimizations, such as fine-grained register management and optimized memory access patterns, that are essential for efficient low-precision computations. In this paper, we introduce Tilus, a domain-specific language designed for General-Purpose GPU (GPGPU) computing that supports low-precision data types with arbitrary bit widths from 1 to 8 while maintaining GPU programmability. Tilus features a thread-block-level programming model, a hierarchical memory space, a novel algebraic layout system, and extensive support for diverse low-precision data types. Tilus programs are compiled into highly efficient GPU programs through automatic vectorization and instruction selection. Extensive experiments demonstrate that Tilus efficiently supports a full spectrum of low-precision data types, and outperforms state-of-the-art low-precision kernels. Compared to existing compilers such as Triton and Ladder, as well as hand-optimized kernels such as QuantLLM and Marlin, Tilus achieves performance improvements of: $1.75\times$, $2.61\times$, $1.29\times$ and $1.03\times$, respectively. We open-source Tilus at https://github.com/NVIDIA/tilus.
Cited by 0Open access
Yaoyao Ding, Bohan Hou, Xiao Zhang et al. · 2025
Serving Large Language Models (LLMs) is critical for AI-powered applications, yet it demands substantial computational resources, particularly in memory bandwidth and computational throughput. Low-precision computation has emerged as a key technique to improve efficiency while reducing resource consumption. Existing approaches for generating low-precision kernels are limited to weight bit widths that are powers of two and suffer from suboptimal performance because of high-level GPU programming abstractions. These abstractions restrict critical optimizations, such as fine-grained register management and optimized memory access patterns, that are essential for efficient low-precision computations. In this paper, we introduce Tilus, a domain-specific language designed for General-Purpose GPU (GPGPU) computing that supports low-precision data types with arbitrary bit widths from 1 to 8 while maintaining GPU programmability. Tilus features a thread-block-level programming model, a hierarchical memory space, a novel algebraic layout system, and extensive support for diverse low-precision data types. Tilus programs are compiled into highly efficient GPU programs through automatic vectorization and instruction selection. Extensive experiments demonstrate that Tilus efficiently supports a full spectrum of low-precision data types, and outperforms state-of-the-art low-precision kernels. Compared to existing compilers such as Triton and Ladder, as well as hand-optimized kernels such as QuantLLM and Marlin, Tilus achieves performance improvements of: 1.75x, 2.61x, 1.29x and 1.03x, respectively. We open-source Tilus at https://github.com/NVIDIA/tilus.
Cited by 0Open access
Matej Adamec, Michal Turčaník · 2023
Can GPT models, specifically those designed for natural language processing, outperform convolutional neural networks (CNN) when it comes to signature detection? Signature detection is one of the most important tasks involved in deep learning. It is used to identify objects and patterns in data. On the other hand, the extremely rapid development of chat bots like chat GPT creates the potential for their use in malware detection with preprocessed code. In this study, the authors compare the performance of signature detection using low-level on the ResNet50 network and high-level programming languages on the GPT 2 model. The results show that signature detection in high-level languages perform better than ResNet50 in low-level languages. This suggests that GPT 2 and general chat bot related models are more suited for signature detection tasks.
Cited by 1
Ding, Yaoyao · Zenodo (CERN European Organization for Nuclear Research) · 2025
This is the artifact for the research paper "Tilus: A Tile-Level GPGPU Programming Language for Low-Precision Computation" in ASPLOS 2026. Version 2: Add script to reproduce the experiment of figure 12
Cited by 0Open access
Wim Vanderbauwhede, Martin Margala, Sai Rahul Chalamalasetti et al. · ENLIGHTEN (Jurnal Bimbingan dan Konseling Islam) · 2009
We present the architecture and programming model for MORA, a coarse-grained reconfigurable processor aimed at multimedia applications. The MORA architecure is a MIMD machine consisting of a 2-D array of reconfigurable cells (RC) with a flexible reconfigurable interconnect network. MORA is designed to support high-throughput data-parallel pipelined processing. We describe the design and implementation of the RC and present simulation results. Apart from its performance, the distinguishing feature of MORA is its design support for programmability. We present a novel assembly language, MORA assembly, which provides a low-level but flexible programming model for the architecture.
Cited by 6
Ding, Yaoyao · Zenodo (CERN European Organization for Nuclear Research) · 2025
This is the artifact for the research paper "Tilus: A Tile-Level GPGPU Programming Language for Low-Precision Computation" in ASPLOS 2026.
Cited by 0Open access
Deva, Gregoryan Billy Hananda · Universitas Kristen Satya Wacana Institutional Repository (Universitas Kristen Satya Wacana) · 2013
Multimedia adalah suatu cara yang digunakan untuk mempresentasikan informasi agar lebih menarik dan mudah dimengerti. Multimedia juga menjadi sangat penting dalam teknologi informasi karena user butuh sesuatu yang menarik dan mudah digunakan. Sekarang ini teknologi sudah berkembang sangat pesat dengan adanya internet yang sangat mudah diakses oleh berbagai kalangan masyarakat. Hal ini dapat dimanfaatkan untuk membuat sebuah presentasi multimedia berbasis web yang dibutuhkan agar penyebaran informasi dapat dilakukan dengan cepat tanpa bergantung waktu dan tempat. Presentasi multimedia berbasis web membutuhkan manipulasi multimedia di dalamnya yang biasa dilakukan dengan menggunakan Synchronized Multimedia Integration Language (SMIL). SMIL merupakan bahasa pemrograman tingkat rendah yang umumnya sulit dipahami. Berdasarkan hal tersebut maka dirancang sebuah library web yang berisi fungsi-fungsi manipulasi multimedia untuk memudahkan programmer dalam memanipulasi multimedia yang akan dipresentasikan ke dalam konten web yang dibangunnya hanya dengan memanggil fungsi-fungsi yang sudah disediakan dalam library web.
Cited by 0Open access
Microsoft Research · Bulletin of Miscellaneous Information (Royal Gardens Kew) · 2005
New ideas in programming languages are best evaluated experimentally. But experimental evaluation is helpful only if there is an implementation that is efficient enough to encourage programmers to use the new features. Ideally, language researchers would build efficient implementations by reusing existing infrastructure, but existing infrastructures do not serve researchers well: in high-level infrastructures, many high-level features are built in and can't be changed, and in low-level infrastructures, it is hard to support important *run-time* services such as garbage collection, exceptions, and so on. I am proposing a different approach: for reuse with many languages, an infrastructure needs *two* low-level interfaces: a compile-time interface and a *run-time* interface. If both interfaces provide appropriate mechanisms, the mechanisms can be composed to build many high-level abstractions, leaving the semantics and cost model up to the client. In this talk, I will illustrate these ideas with examples drawn from two parts of the C-- language infrastructure: exception dispatch and procedure calls. I will focus on the mechanisms that make it possible for you to choose the semantics and cost model you want. For exceptions, these mechanisms are drawn from both compile-time and run-time interfaces, and together they enable you to duplicate all the established techniques for implementing exceptions. For procedure calls, the mechanisms are quite different; rather than provide low-level mechanisms that combine to form different kinds of procedure calls, I have found it necessary to extend the compile-time interface to enable direct control of the semantics and cost of procedure calls. I will also sketch some important unsolved problems regarding mechanisms to support advanced control features such as threads and first-class continuations. ©2005 Microsoft Corporation. All rights reserved.
Cited by 0
Yuan Zhou, Bradley Gram-Hansen, Tobias Kohn et al. · arXiv (Cornell University) · 2019
We develop a new Low-level, First-order Probabilistic Programming Language (LF-PPL) suited for models containing a mix of continuous, discrete, and/or piecewise-continuous variables. The key success of this language and its compilation scheme is in its ability to automatically distinguish parameters the density function is discontinuous with respect to, while further providing runtime checks for boundary crossings. This enables the introduction of new inference engines that are able to exploit gradient information, while remaining efficient for models which are not everywhere differentiable. We demonstrate this ability by incorporating a discontinuous Hamiltonian Monte Carlo (DHMC) inference engine that is able to deliver automated and efficient inference for non-differentiable models. Our system is backed up by a mathematical formalism that ensures that any model expressed in this language has a density with measure zero discontinuities to maintain the validity of the inference engine.
Cited by 10Open access
Gram-Hansen, BJ, Kohn, T, Wood, F et al. · Oxford University Research Archive (ORA) (University of Oxford)
We develop a new Low-level, First-order Probabilistic Programming Language (LF-PPL) suited for models containing a mix of continuous, discrete, and/or piecewise-continuous variables. The key success of this language and its compilation scheme is in its ability to automatically distinguish parameters the density function is discontinuous with respect to, while further providing runtime checks for boundary crossings. This enables the introduction of new inference engines that are able to exploit gradient information, while remaining efficient for models which are not everywhere differentiable. We demonstrate this ability by incorporating a discontinuous Hamiltonian Monte Carlo (DHMC) inference engine that is able to deliver automated and efficient inference for non-differentiable models. Our system is backed up by a mathematical formalism that ensures that any model expressed in this language has a density with measure zero discontinuities to maintain the validity of the inference engine.
Cited by 0Open access
Bagge Carlson, Fredrik, Haage, Mathias · Lund University Publications (Lund University) · 2017
A generic robot library written in the Julia programming language is presented. A bridging framework, also presented, exposes the ABB externally guided motion research interface (EGMRI) low-level robot motion correction interface to the Julia language and other entities, such as Python and ROS. A usage example on Julia contact force estimation on an ABB YuMi robot concludes the document.
Cited by 0Open access
Dimosthenis E. Bolanakis, G.A. Evangelakis, Euripidis Glavas et al. · Computer Applications in Engineering Education · 2009
Abstract Assembly language programming has been for long used as a teaching tool for computer systems. However, this approach suffers from the difficulties that are inherent to its low‐level instructions. The present teaching approach merges the assembly language course with microcomputers technology by means of structured pseudo‐coding. The method is based on a classification of the assembly instructions in a novel inventory, which is subsequently used for the composition of the basic high‐level programming possibilities, thus overcoming the limited assembly abilities. The proposed methodology aims at helping the students to make the parallelization between the assembly language programming for 8‐bit microcontrollers and a higher level programming. © 2009 Wiley Periodicals, Inc. Comput Appl Eng Educ 19: 525–537, 2011
Cited by 16
Martin Nilsson Sics · 1994
Introduction Programming autonomous robots requires a language that is able to express both very high-level algorithms as well as low-level control. We are developing a concurrent programming framework, which is able to handle both complex, high-level algorithms and very low-level hardware control in the same language. In order to illustrate these features we shall describe briefly the approach to implementing one high-level algorithm, which is constraint solving. As an example of low-level programming, we shall outline how behaviours can be programmed using the same programming concepts. We assume familiarity with symbolic languages such as Lisp, Prolog, the Behavior Language, and some acquaintance with constraint solving. 2 The basic programming framework The basic programming language we use, Real-Time Fleng, is a concurrent programming language derived from symbolic languages such as Lisp, Prolog, and Guarded Horn Clauses. It has an inhere
Cited by 0
Amlal El Mahrouss · Zenodo (CERN European Organization for Nuclear Research) · 2026
Design of the Nectar Programming Language. Nectar as presented in its primer is a systems compiled programming language—designed for low-level systems programming with high-level abstractions.
Cited by 0Open access
Amlal El Mahrouss · Zenodo (CERN European Organization for Nuclear Research) · 2026
Design of the Nectar Programming Language. Abstract: Nectar as presented in its primer is a systems compiled programming language—designed for low-level systems programming with high-level abstractions.
Cited by 0Open access
Amlal El Mahrouss · Zenodo (CERN European Organization for Nuclear Research) · 2026
Design of the Nectar Programming Language. Abstract: Nectar as presented in its primer is a systems compiled programming language—designed for low-level systems programming with high-level abstractions.
Cited by 0Open access