机器人论文阅读
Vic_Hao
这个作者很懒,什么都没留下…
展开
-
Control of a Nonholonomic Mobile Robot Using Neural Networks
Introduction这篇文章主要就是提出了一个control structure that makes possible the integration of a kinematic controller and a neural network computed-torque controller for nonholonomic mobile robots.Mobile robot n...原创 2019-07-08 13:48:48 · 346 阅读 · 1 评论 -
Slip‐compensated odometry for tracked vehicle on loose and weak slope
Introductiontrack和地面打滑的时候,odometry会失效。因此就要估计track和地面的打滑,因此就要建立一个estimating the interaction between a vehicle and the ground的model, 但是通常这些model都有terrain-specific的参数,所以很难应用在unknown environments. 在这个研究...原创 2019-07-08 13:49:22 · 318 阅读 · 0 评论 -
Path Tracking Control of Tracked Vehicles
Introduction这篇文章主要就是建立了一个dynamic model (including track-soil interactions), 然后建立了一个controller。然后在仿真器中测试了一下。The result is an accurate control of the lateral offset and forward velocity error along a ...原创 2019-07-08 14:21:30 · 592 阅读 · 0 评论 -
Robust Constrained Learning-based NMPC enabling reliable mobile robot path tracking
Introduction这篇文章的目的就是achieving robust constrained, high performance path-tracking in spite of unknown disturbances.这篇文章的思路:simple process model, high model uncertainty →learn\rightarrow^{learn}→lear...原创 2019-06-24 23:24:25 · 455 阅读 · 1 评论 -
Experimental Analysis of Models for Trajectory Generation on Tracked Vehicles
Introduction这篇文章提示了一下feedback的本质作用:avert the issue of unmodeled dynamics,但是通常会有三方面的问题:There is no obvious error signal available at the rate and scale of a single, rapid maneuver: GPS errors are t...原创 2019-06-28 16:52:13 · 299 阅读 · 0 评论 -
Motion Control of Tracked Vehicle based on Contact Force Model
introductionThis paper propose an navigation method for tracked vehicles based on a contact force model that enables the robot to compensate for collisions with obstacles.这篇文章的意思是collision是不能避免的due ...原创 2019-07-04 11:56:18 · 752 阅读 · 0 评论 -
Feedback stabilizer-based trajectory planning of mobile robots with kinematic constraints
IntroductionContributionThey propose a generic approach which first uses an (arbitrary) feedback stabilizer to generate the ‘path’ and then rebuilt the corresponding ‘trajectory’ along this ‘path’ t...原创 2019-07-11 14:57:25 · 351 阅读 · 0 评论 -
Path Following Control for Tracked Vehicles Based on Slip-Compensating Odometry
IntroductionIn this paper, we propose a new odometry method to improve the accuracy. One of the features of the proposed method is that it can be achieved by a gyro-sensor, encoders (which detect the...原创 2019-07-15 20:55:50 · 315 阅读 · 0 评论 -
Terrain Adaptive Odometry for Mobile skid-steer Robots
IntroductionThis paper proposes a novel approach improving precision and reliability of odometry of skid-steer mobile robots by means inspired by robotic terrain classification (RTC).NoveltyIn cont...原创 2019-07-15 20:56:07 · 320 阅读 · 0 评论 -
Robust Sampling-based Motion Planning for Autonomous Tracked Vehicles in Deformable High Slip Terrai
preface首先要点一下motion planning algorithm的作用:The motion planning algorithm should find the optimal and probabilistically feasible trajectory from the initial state zinitz_{init}zinit to the goal regio...原创 2019-07-11 17:03:23 · 194 阅读 · 0 评论 -
Adaptive trajectory control of off-road mobile robots: A multi-model observer approach
Introductionoff-road conditionModelingA combined model is obtained and used for the purpose of determining the parallel kinematics and dynamics of the robot.Previous Estimation of Sideslip Angles ...原创 2019-07-11 23:24:06 · 164 阅读 · 0 评论 -
The ETH-MAV Team in the MBZ International Robotics Challenge
Introduction这篇文章介绍了ETH MAV team 参加2017年MBZIRC搭建的MAV system platform (including hardware and software)先简要介绍一下MBZIRC的任务:The competition consisted of three individual challenges and a triathlon-type G...原创 2019-08-03 22:43:43 · 283 阅读 · 0 评论 -
Nonlinear Systems Identification Using Deep Dynamic Neural Networks
IntroductionThis paper investigates the effectiveness of deep neural networks in the modeling of dynamical systems with complex behavior.Three deep neural network structures are trained on sequentia...原创 2019-07-28 21:31:02 · 264 阅读 · 0 评论 -
Deep Learning Helicopter Dynamics Models
AbstractThis paper pose the dynamics modeling problem as direct high-dimensional regression, and take inspiration from recent results in Deep Learning to represent the helicopter dynamics with a ReLU...原创 2019-09-09 19:59:49 · 460 阅读 · 0 评论 -
Online Utility-Optimal Trajectory Design for Time-Varying Ocean Environments
IntroductionTask:energy-efficienttime-varyingcurrentgoalMethodformulate the general utility-optimal online trajectory learning problemThe trajectory learning problem is expressed as that ...原创 2019-09-26 13:32:08 · 177 阅读 · 0 评论 -
Learning to See the Wood for the Trees: Deep Laser Localization in Urban and Natural Environments on
Topiclaser-based localization in both urban and natural environments, which is suitable for online applications这篇文章提出了一种基于深度学习的方法,通过对比triplets(anchor, positive, negative examples),可以直接从3D point clou...原创 2019-06-22 14:54:32 · 173 阅读 · 0 评论 -
Approximating Kinematics for Tracked Mobile Robots
Idea在运动平面上的瞬时旋转中心是相对于robot是dynamics-dependent,but lie within a bounded area. Thus, optimizing constant ICR positions for a particular terrain results in an approximate kinematic model for tracked mob...原创 2019-06-22 14:14:58 · 418 阅读 · 0 评论 -
Tracked Mobile Robot Control: Hybrid Approach
ComplexityThe kinematics of a tracked vehicle is very complex, since:infinite number of contact points between the track and the terraininfinite closed-link chains which make slippage unavoidable...原创 2019-06-20 22:06:28 · 206 阅读 · 0 评论 -
Probabilistic Foot Contact Estimation by Fusing Information from Dynamics and Differential/Forward K
Goal:measure force without force sensor这类topic的文章要关注做的是collision detection还是contact detectioncollision detection: detect an impactcontact detection: detect accurate timing of the loss of the conta...原创 2019-02-26 12:41:45 · 375 阅读 · 0 评论 -
Robust Recovery Controller for a Quadrupedal Robot using Deep Reinforcement Learning
Goalthe ability to recover from a fall of a legged robotContributiona hierarchical behavior-based controllercontroller有4个neural network policies, 包括3个behaviors和1个behavior selector.muti-task DRL...原创 2019-02-26 20:14:32 · 296 阅读 · 0 评论 -
Control of a Quadrotor with Reinforcement Learning
GoalControl a quadrotor with a neural network trained using reinforcement learning.Policy network is a function directly mapping a state to rotor thrusts.Related WorkGuided Policy Search with a MP...原创 2019-03-08 14:23:50 · 479 阅读 · 0 评论 -
Automous Identification and Goal-directed Invocation of Event-Predictive Behavioral Primitives
MotivationThe problem of learning meaningful, compositional abstractions from sensorimotor experiences remains an open challenge. A large challenge for the brain as well as artificial cognitive syste...原创 2019-03-04 16:57:58 · 268 阅读 · 0 评论 -
Direct State-to-Action Mapping for High DOF Robots Using ELM
GoalRelated WorkTo generate a general state-to-action mapping for a high dimensional system, optimizing the mapping using ‘trial and error’ basis has limitations due to the fact that ...原创 2019-03-07 13:50:55 · 136 阅读 · 0 评论 -
An Application of Reinforcement Learning to Aerovbatic Helicopter Flight
这是早期成功应用RL到robotcis的文章。GoalChallenge自动直升机是一个具有挑战性的控制问题,具有high-dimensional, asymmetric, noisy, nonlinear, non-minimum phase dynamics的特点。Contribution1.Learning a Helicopter Model from Flight DataC...原创 2019-03-07 15:08:04 · 107 阅读 · 0 评论 -
A straight-line motion tracking control of hydraulic excavator system
GoalThe resulting control law has been applied to straight-line motions of a 13 ton hydraulic excavator with a bucket speed of 0.5 m/s, a speed level at which skilful operators work.ChallengeStraig...原创 2019-03-14 20:30:43 · 170 阅读 · 0 评论 -
Exploration and Apprenticeship Learning in Reinforcement Learning
Goal本文提出apprenticeship learning, in which a teacher demonstration of the task is available。given theinitial demonstration, no explicit exploration is necessary, and we can attain near-optim...原创 2019-03-07 19:57:10 · 269 阅读 · 0 评论 -
Constructing Hydraulic Robot Models Using Memory-Based Learning
Goalconstruct a complete excavator machine model consists of an actuator model and a linkage dynamicsChallenge and Related WorkContributionpropose a method for constructing hydraulic machine model...原创 2019-03-25 23:04:04 · 120 阅读 · 0 评论 -
Neural Network Dynamics for Model-Based Deep Reinforcement Learniing with Model-Free Fine-Tuning
Goal怎样在model-based reinforcement learning中使用neural-network创建的system dynamics怎样使用model-based reinforcement learning来加速model-free reinforcement learningRelated WorkThe most efficient model-based a...原创 2019-03-26 17:21:17 · 317 阅读 · 0 评论 -
Dynamics Movement Primitives-A Framework for Motor Control in Humans and Humanoid Robotics
Research Topic基本假设:movement sequences consist of sequence segments, executed either in sequence or with partial or completely overlap.identifying movement primitives (a.k.a. units of actions, basis ...原创 2019-04-15 23:18:47 · 733 阅读 · 0 评论 -
Robotic Pick-and-Place of Novel Objects in Clutter with Multi-Affordance Grasping...
Research TopicA robotic pick-and-place system that is capable of grasping and recognizing both known and novel objects in cluttered environments.这里的novel object是指不需要task-specific training data for n...原创 2019-04-16 23:13:04 · 1016 阅读 · 0 评论 -
Multi-resolution Planning for Earthmoving
Research Topica multi-resolution planning and execution scheme:At the highest level (coarse planner)is a coarse planning scheme that uses geometry of the site and the goal configuration of the ter...原创 2019-04-22 22:11:00 · 153 阅读 · 0 评论 -
See, feel, act: Hierarchical learning for complex manipulation skills with multisensory fusion
Research TopicInspired by:Humans are able to seamlessly integrate tactile and visual stimuli with their intuitions to explore and execute complex manipulation skills.原创 2019-04-28 09:49:08 · 481 阅读 · 0 评论 -
Autonomous Robotic Stone Stacking with Online next Best Object Target Pose Planning
Goal: 用不规则形状的石头(已知的set,也有pre-scanned model)搭建一个vertical towerContribution:pose searching algorithm(stacking algorithm)目的是在object set中找到best object 和它的target pose。search space: discrete object spa...原创 2019-02-23 21:27:40 · 167 阅读 · 0 评论