Camera relocalization is pivotal in computer vision, with applications in AR, drones, robotics, and autonomous driving. It estimates 3D camera position and orientation (6-DoF) from images. Unlike traditional methods l...
详细信息
ISBN:
(纸本)9798350384581;9798350384574
Camera relocalization is pivotal in computer vision, with applications in AR, drones, robotics, and autonomous driving. It estimates 3D camera position and orientation (6-DoF) from images. Unlike traditional methods like SLAM, recent strides use deep learning for direct end-to-end pose estimation. We propose EffLoc, a novel efficient Vision Transformer for single-image camera relocalization. EffLoc's hierarchical layout, memory-bound self-attention, and feed-forward layers boost memory efficiency and inter-channel communication. Our introduced sequential group attention (SGA) module enhances computational efficiency by diversifying input features, reducing redundancy, and expanding model capacity. EffLoc excels in efficiency and accuracy, outperforming prior methods, such as AtLoc and MapNet. It thrives on large-scale outdoor car-driving scenario, ensuring simplicity, end-to-end trainability, and eliminating handcrafted loss functions.
The energy of ocean waves is the key distinguishing factor of marine environments compared to other aquatic environments such as lakes and rivers. Waves significantly affect the dynamics of marine vehicles;hence it is...
详细信息
ISBN:
(纸本)9781728190778
The energy of ocean waves is the key distinguishing factor of marine environments compared to other aquatic environments such as lakes and rivers. Waves significantly affect the dynamics of marine vehicles;hence it is imperative to consider the dynamics of vehicles in waves when developing efficient control strategies for autonomous surface vehicles (ASVs). However, most marine simulators available open-source either exclude dynamics of vehicles in waves or use methods with high computational overhead. This paper presents ASVLite, a computationally efficient ASV simulator that uses frequency domain analysis for wave force computation. ASVLite is suitable for applications requiring low computational overhead and high run-time performance. Our tests on a Raspberry Pi 2 and a mid-range desktop computer show that the simulator has a high run-time performance to efficiently simulate irregular waves with a component wave count of up to 260 and large-scale swarms of up to 500 ASVs.
A functional system architecture for semi-autonomous land vehicles is presented supporting flexible automation of driving functions. The system architecture incorporates different levels of information processing, whi...
详细信息
ISBN:
(纸本)0780344235
A functional system architecture for semi-autonomous land vehicles is presented supporting flexible automation of driving functions. The system architecture incorporates different levels of information processing, which can be distinguished by their knowledge representation and processing. Concepts enabling flexible automation of driving functions in semiautonomous vehicles are proposed.
Image feature based landmark detection is widely used in machine vision for mobile robotics applications. Advanced image feature extraction algorithms offer robust keypoint detection as being invariant to numerous tra...
详细信息
Interactive driving scenarios, such as lane changes. merges and unprotected turns, are some of the most challenging situations for autonomous driving. Planning in interactive scenarios requires accurately modeling the...
详细信息
ISBN:
(纸本)9781728190778
Interactive driving scenarios, such as lane changes. merges and unprotected turns, are some of the most challenging situations for autonomous driving. Planning in interactive scenarios requires accurately modeling the reactions of other agents to different future actions of the ego agent. We develop end-to-end models for conditional behavior prediction (CBP) that take as an input a query future trajectory for an ego agent, and predict distributions over future trajectories for other agents conditioned on the query. Leveraging such a model, we develop a general-purpose agent interactivity score derived from probabilistic first principles. The interactivity score allows us to find interesting interactive scenarios for training and evaluating behavior prediction models. We further demonstrate that the proposed score is effective for agent prioritization under computational budget constraints.
Neural Architecture Search (NAS) has proved effective in offering outperforming alternatives to handcrafted neural networks. In this paper we analyse the benefits of NAS for image classification tasks under strict com...
详细信息
ISBN:
(纸本)9781728190778
Neural Architecture Search (NAS) has proved effective in offering outperforming alternatives to handcrafted neural networks. In this paper we analyse the benefits of NAS for image classification tasks under strict computational constraints. Our aim is to automate the design of highly efficient deep neural networks, capable of offering fast and accurate predictions and that could be deployed on a low-memory, low-power system-on-chip. The task thus becomes a three-party trade-off between accuracy, computational complexity, and memory requirements. To address this concern, we propose Multi-Scale Resource-Aware Neural Architecture Search (MS-RANAS). We employ a one-shot architecture search approach in order to obtain a reduced search cost and we focus on an anytime prediction setting. Through the usage of multiple-scaled features and early classifiers, we achieved state-of-the-art results in terms of accuracy-speed trade-off.
The present paper discusses an optimal control method of biological robot arm which has redundancy of the mapping from the control input to the task goal. The control input space is divided into a couple of subspaces ...
详细信息
ISBN:
(纸本)0780372727
The present paper discusses an optimal control method of biological robot arm which has redundancy of the mapping from the control input to the task goal. The control input space is divided into a couple of subspaces according to a priority order depending on the progress and stability of learning. In the proposed method, the search noise which is required for reinforcement learning is restricted within the first priority subspace. Then the constraint is relaxed with the progress of learning, and the search space extends to the second priority subspace in accordance with the history of learning. The method was applied to the musculoskeletal system as an example of biological control systems. Dynamic manipulation is obtained through reinforcement learning with no previous knowledge of the arm's dynamics. The effectiveness of the proposed method is shown by computational simulation.
This paper introduces an artificial intelligent technique to solve a complex problem, an integrated production-distribution scheduling (IPDS) problem. The problem governs with capacitated single-facility, multi-produc...
详细信息
ISBN:
(纸本)9781509060887
This paper introduces an artificial intelligent technique to solve a complex problem, an integrated production-distribution scheduling (IPDS) problem. The problem governs with capacitated single-facility, multi-product, multi-retailer, and a homogenous vehicle fleet. The objective is to minimize the total costs of the whole supply chain. We make use of an evolutionary computation technique, particle swarm optimization (PSO), to solve the problem. The proposed method is tested on 7 datasets and compared with a competitive optimizer, CPLEX. The results show that the proposed method yields acceptable solutions with low computational time.
This paper proposes a general-purpose, highly configurable controller architecture for a wide variety of sensor-based programmable automation equipment. When fully implemented, this system will provide a flexible prog...
详细信息
ISBN:
(纸本)0262061015
This paper proposes a general-purpose, highly configurable controller architecture for a wide variety of sensor-based programmable automation equipment. When fully implemented, this system will provide a flexible programming and computational environment to support a wide range of research spanning the development of new functional technology, manipulator motion control methods, explicit programming, model-based program generation, and integrated manufacturing experiments. The system consists of an interactive Programming System connected through a shared memory to a multiple-processor Real Time System that performs time-critical operations. The Programming System executes programs written in an enhanced version of AML and transmits high level commands, called verbs, to the Real Time System for execution. The processes themselves are specified as data flow graphs involving computational functions and shared state vector variables.
The current construction industry has different environmental characteristics compared to other industries. Problems such as the instability of labor force demand and supply and the increasing labor costs are surfacin...
详细信息
The current construction industry has different environmental characteristics compared to other industries. Problems such as the instability of labor force demand and supply and the increasing labor costs are surfacing. Thus, in response to a high growth rate of the construction market, it is advisable to lower the level of labor force dependence and increase efficiency by applying a specialized automation in a construction site for improving its productivity. Currently, a construction automation robot is being developed mainly as a professional service robot. The majority of its abilities rely on the human's control skill and its level of recognizing the external environment is also inadequate. It is therefore inappropriate to say that the complete and fully automation in construction is accomplished. A high level of environmental cognitive skill and such as the same level of human ability in judging the situation are needed at construction sites in order to accomplish complete automation. It is expected that in the future, the automation and robotics in construction would be replaced with robots having artificial intelligence similar to that of humans. However, the current level of technology cannot meet the specifications yet, it is appropriate to utilize the Human-Robot Cooperation (HRC) technology as an alternative to the current and future vision of robotics and automation in construction. The HRC allows a high efficiency at the present automation level because it uses both advantages of humans who have a remarkable set of nervous system and of robots which functions at a superior physical ability.
暂无评论