Model based vehicle detection and tracking for autonomous urban driving

Situational awareness is crucial for autonomous driving in urban environments. This paper describes the moving vehicle detection and tracking module that we developed for our autonomous driving robot Junior. The robot won second place in the Urban Grand Challenge, an autonomous driving race organized by the U.S. Government in 2007. The module provides reliable detection and tracking of moving vehicles from a high-speed moving platform using laser range finders. Our approach models both dynamic and geometric properties of the tracked vehicles and estimates them using a single Bayes filter per vehicle. We present the notion of motion evidence, which allows us to overcome the low signal-to-noise ratio that arises during rapid detection of moving vehicles in noisy urban environments. Furthermore, we show how to build consistent and efficient 2D representations out of 3D range data and how to detect poorly visible black vehicles. Experimental validation includes the most challenging conditions presented at the Urban Grand Challenge as well as other urban settings.     

Reactive navigation in dynamic environment using a multisensorpredictor

A reactive navigation system for an autonomous mobile robot in unstructured dynamic environments is presented. The motion of moving obstacles is estimated for robot motion planning and obstacle avoidance. A multisensor-based obstacle predictor is utilized to obtain obstacle-motion information. Sensory data from a CCD camera and multiple ultrasonic range finders are combined to predict obstacle positions at the next sampling instant. A neural network, which is trained off-line, provides the desired prediction on-line in real time. The predicted obstacle configuration is employed by the proposed virtual force based navigation method to prevent collision with moving obstacles. Simulation results are presented to verify the effectiveness of the proposed navigation system in an environment with multiple mobile robots or moving objects. This system was implemented and tested on an experimental mobile robot at our laboratory. Navigation results in real environment are presented and analyzed.     

基于主动视觉和超声的机器人目标跟踪系统

运动目标跟踪技术是未知环境下移动机器人研究领域的一个重要研究方向。该文提出了一种基于主动视觉和超声信息的移动机器人运动目标跟踪设计方法,利用一台SONY EV-D31彩色摄像机、自主研制的摄像机控制模块、图像采集与处理单元等构建了主动视觉系统。移动机器人采用了基于行为的分布式控制体系结构,利用主动视觉锁定运动目标,通过超声系统感知外部环境信息,能在未知的、动态的、非结构化复杂环境中可靠地跟踪运动目标。实验表明机器人具有较高的鲁棒性,运动目标跟踪系统运行可靠。     

基于双计算机的仿人机器人的视觉跟踪系统

运动目标的实时跟踪是机器人视觉的关键技术之一。设计了仿人机器人的视觉跟踪系统,系统采用双计算机,分别负责视觉信息的处理和运动单元的控制,两台计算机通过Memolink进行通讯。基于Windows的视觉信息处理子系统实现运动目标的分割,状态估计和预测。运动控制子系统采用RTlinux实时操作系统,利用PD控制器控制关节运动。实验验证了系统的稳定性和实时性。     

A Probabilistic Framework for Sonar Scan Matching Localization

Scan matching is a popular localization technique based on comparing two sets of range readings gathered at consecutive robot poses. Scan matching algorithms implicitly assume that matching readings correspond to the same object in the environment. This is a reasonable assumption when using accurate sensors such as laser range finders and that is why they are extensively used to perform scan matching localization. However, when using other sensors such as ultrasonic range finders or visual sonar, this assumption is no longer valid because of their lower angular resolution and the sparsity of the readings. In this paper we present a sonar scan matching framework, the spIC, which is able to deal with the sparseness and low angular resolution of sonar sensors. To deal with sparseness, a process to group sonar readings gathered along short robot trajectories is presented. Probabilistic models of ultrasonic and odometric sensors are defined to cope with the low sonar angular resolution. Consequently, a probabilistic scan matching process is performed. Finally, the correction of the whole robot trajectory involved in the matching process is presented as a constrained optimization problem.     

基于激光雷达的移动机器人动态障碍跟踪

动态环境下运动物体跟踪是移动机器人研究的难点之一;文章提出了一种基于激光雷达的自主动态障碍检测与跟踪方法;该方法首先利用最近邻聚类法将环境数据聚类为不同的障碍物;然后利用最近邻特征匹配算法关联相邻两帧的障碍物;最后提出一种新的基于障碍物时空关联性分析的的障碍物动静态识别算法,并采用α-β滤波算法对动态障碍的位置和速度进行了估计;利用机器人平台对该方法进行验证,实验结果表明了其有效性。     

移动机器人运动目标跟踪系统设计

利用SONYEV-D31摄像机和自主研发的摄像机控制模块,构建了一套主动视觉子系统,并将该子系统应用于RIRA-Ⅱ型移动机器人上,实现了移动机器人运动目标自动跟踪功能。RIRA-Ⅱ移动机器人采用了由一组分布式行为模块和集中命令仲裁器组成的基于行为的分布式控制体系结构。各行为模块基于领域知识通过反应方式产生投票,由仲裁器产生动作指令,机器人完成相应的动作。在设置了障碍、窄通道以及模拟墙体的复杂环境下进行运动目标跟踪实验,实验表明运动目标跟踪系统运行可靠,具有较高的鲁棒性。     

Optimal Representative Blocks for the Efficient Tracking of a Moving Object

Optimal representative blocks are proposed for an efficient tracking of a moving object and it is verified experimentally by using a mobile robot with a pan‐tilt camera. The key idea comes from the fact that when the image size of a moving object is shrunk in an image frame according to the distance between the camera of mobile robot and the moving object, the tracking performance of a moving object can be improved by shrinking the size of representative blocks according to the object image size. Motion estimation using edge detection (ED) and block‐matching algorithm (BMA) are often used in the case of moving object tracking by vision sensors. However, these methods often miss the real‐time vision data since these schemes suffer from the heavy computational load. To overcome this problem and to improve the tracking performance, the optimal representative block that can reduce a lot of data to be computed is defined and optimized by changing the size of the representative block according to the size of object in the image frame. The proposed algorithm is verified experimentally by using a mobile robot with a two degree‐of‐freedom active camera. © 2004 Wiley Periodicals, Inc.     

六足机器人动态目标检测与跟踪系统研究

动态目标检测与目标跟踪是图像领域的热点研究问题,为研究其在移动机器人领域的应用价值,设计了六足机器人动态目标检测与跟踪系统。针对非刚体运动目标容易被检测为多个分散区域的问题提出区域合并算法,并通过对称匹配、自适应外点滤除对运动背景进行精确补偿,最终基于背景补偿法实现对运动目标的精确检测。研究了基于KCF（Kernel Correlation Filter）的目标跟踪算法在六足机器人平台上的应用,设计了自适应跟踪算法实现六足机器人对运动目标的角度跟踪。将运动目标检测及跟踪算法应用于六足机器人系统。实验表明,在六足机器人移动过程中,系统可对运动目标进行精确检测与跟踪。     

The Tracking of a Moving Object by a Mobile Robot Following the Object’s Sound

The tracking of a moving object with a mobile robot has been implemented based on the detected sound from the moving object using a microphone array. The difference between the travel times of the sound source to each of the three microphones mounted to the robot has been used to calculate the distance and orientation of the sound source. The cross-correlations between the received signals have been used to detect the individual sound signal from the object and to calculate the time difference between two signals. This provides reliable and precise time differences among the sound signals arrived at the microphones compared to the conventional method. In order to determine the tracking direction to the sound source, Fuzzy rules have been applied; the results are used for real-time control of the mobile robot. The efficiency of the proposed algorithm has been demonstrated through real-world experiments and compared to the conventional approach.     

People detection through quantified fuzzy temporal rules

The knowledge about the position and movement of people is of great importance in mobile robotics for implementing tasks such as navigation, mapping, localization, or human-robot interaction. This knowledge enhances the robustness, reliability and performance of the robot control architecture. In this paper, a pattern classifier system for the detection of people using laser range finders data is presented. The approach is based on the quantified fuzzy temporal rules (QFTRs) knowledge representation and reasoning paradigm, that is able to analyze the spatio-temporal patterns that are associated to people. The pattern classifier system is a knowledge base made up of QFTRs that were learned with an evolutionary algorithm based on the cooperative-competitive approach together with token competition. A deep experimental study with a Pioneer II robot involving a five-fold cross-validation and several runs of the genetic algorithm has been done, showing a classification rate over 80%. Moreover, the characteristics of the tests represent complex and realistic conditions (people moving in groups, the robot moving in part of the experiments, and the existence of static and moving people).     

Optimal rendezvous-point selection for robotic interception ofmoving objects

A number of active prediction planning and execution (APPE) systems have recently been proposed for robotic interception of moving objects. The cornerstone of such systems is the selection of a robot-object rendezvous-point on the predicted object trajectory. Unlike tracking-based systems, which minimize the state difference between the object and the robot at each control period, in this methodology the robot is sent directly to the selected rendezvous-point. A fine-motion tracking strategy would then be employed for grasping the moving object. Herein, a novel strategy for selecting the optimal (earliest) rendezvous-point is presented. For objects with predictable trajectories, this is a significant improvement over previous APPE strategies which select the rendezvous-point from a limited number of non-optimally chosen candidates.     

Navigation Using an Environmental Magnetic Field for Outdoor Autonomous Mobile Robots

GPS and laser range finders are generally utilized in current robot navigation. However, information from the magnetic field and electronic compass is not, since it is dynamically changing at every position. In this paper, the relationship between the intensity of a magnetic field in the environment and its position is taken into account by utilizing a three-axis magnetic sensor to scan the magnetic field in the environment to build a database. The mobile robot navigates by performing trajectory tracking based on the database. The experimental results show that by applying the proposed method, the mobile robot is able to navigate in an outdoor environment with reliable accuracy.     

Navigation system including associative memory

A new adaptive linear robot control system for a robot work cell that can visually track and intercept stationary and moving objects undergoing arbitrary motion anywhere along its predicted trajectory within the robot's workspace is presented in this paper. The proposed system was designed by integrating a stationary monocular CCD camera with off-the-shelf frame grabber and an industrial robot operation into a single application on the MATLAB platform. A combination of the model based object recognition technique and a learning vector quantization network is used for classifying stationary objects without overlapping. The optical flow technique and the MADALINE network are used for determining the target trajectory and generating the predicted robot trajectory based on visual servoing, respectively. The necessity of determining a model of the robot, camera, all the stationary and moving objects, and environment is eliminated. The location and image features of these objects need not be preprogrammed, marked and known before, and any change in a task is possible without changing the robot program. After the learning process on the robot, it is shown that the KUKA robot is capable of tracking and intercepting both stationary and moving objects at an optimal rendezvous point on the conveyor accurately in real-time.     

基于运动区域检测的运动目标跟踪算法*

针对传统基于模板匹配的运动目标跟踪算法存在着计算量大、模板漂移导致跟踪失败的问题,提出了一种基于运动区域检测的运动目标跟踪算法。该算法通过采用光流法对目标运动区域进行估计,计算出光流场区域的形心,确定待匹配图相匹配范围,再用模板框在已确定区域进行模板匹配跟踪。根据某开放实验室行人录像跟踪实验表明,本算法能够有效解决模板漂移问题,提高了跟踪实时性, 实现了视频对象目标的跟踪。     

Virtual target tracking of mobile robot and its application to formation control

A virtual target tracking approach is proposed for kinematic control of mobile robot. In the controller, linear and angular velocity inputs are generated by using the local data of robot position and orientation along with the estimated velocity of target object. Applying the proposed approach to a cooperative robot group with arbitrary number of multiple mobile robots, it is possible to create various robot formations for cooperative navigation and tracking of moving object. The developed controller is shown to be stable and convergent through theoretical proof and a series of experiments.     

基于激光传感器移动机器人的目标定位和多模式跟踪

移动机器人对运动目标的感知和跟踪是实现机器人与环境交互的一项重要能力。针对移动机器人以人为目标的跟踪中在复杂动态环境下经常出现的目标丢失和跟踪模式单一的问题,提出了基于机器学习的人物目标识别算法。该算法可以处理复杂环境下的目标检测和定位。同时设计了交互多模型跟踪算法,可以较好的跟踪以不规律模式运动的目标。最后在交龙移动机器人平台上实现了整个系统,验证了人物目标检测和多模式跟踪算法的鲁棒性和优越性。     

Adaptive neural controller for cooperative multiple robot manipulator system manipulating a single rigid object

In this article, an adaptive neural controller is developed for cooperative multiple robot manipulator system carrying and manipulating a common rigid object. In coordinated manipulation of a single object using multiple robot manipulators simultaneous control of the object motion and the internal force exerted by manipulators on the object is required. Firstly, an integrated dynamic model of the manipulators and the object is derived in terms of object position and orientation as the states of the derived model. Based on this model, a controller is proposed that achieves required trajectory tracking of the object as well as tracking of the desired internal forces arising in the system. A feedforward neural network is employed to learn the unknown dynamics of robot manipulators and the object. It is shown that the neural network can cope with the unknown nonlinearities through the adaptive learning process and requires no preliminary offline learning. The adaptive learning algorithm is derived from Lyapunov stability analysis so that both error convergence and tracking stability are guaranteed in the closed loop system. Finally, simulation studies and analysis are carried out for two three-link planar manipulators moving a circular disc on specified trajectory.     

Multisensor controlled robot system for recognizing and tracking moving multiple objects

This paper presents a cost-efficient, real-time vision-sensor system for identifying, locating and tracking objects that are unknown and randomly placed on a moving conveyor belt. The visual information obtained from a conventional frame-store unit and an end-effector based proximity sensor outputs are incorporated in a fuzzy-logic control algorithm to make the robotic manipulator grasp moving objects. The robot movements are going to be the result of the comparative measurements made by the sensors after the motion of the moving target is predicted and the gripper is brought into a zone close to the object to be grasped by the application of a vision system. The mobile object is traced by controlling the motion of the end-effector with an end-effector based infrared proximity sensors and conveyor position encoder by keeping the gripper's axis to pass through a median plane of the moving object. With this procedure and using the fuzzy-logic control, the system is adapted to pursue of a mobile object. Laboratory experiments are presented to demonstrate the performance of this system. ©1999 John Wiley & Sons, Inc.     

Optimal Motion Planning of Robotic Manipulators Removing Mobile Objects Grasped in Motion

The following study deals with motion optimization of robot arms having to transfer mobile objects grasped when moving. This approach is aimed at performing repetitive transfer tasks at a rapid rate without interrupting the dynamics of both the manipulator and the moving object. The junction location of the robot gripper with the object, together with grasp conditions, are partly defined by a set of local constraints. Thus, optimizing the robot motion in the approach phase of the transfer task leads to the statement of an optimal junction problem between the robot and the moving object. This optimal control problem is characterized by constrained final state and unknown traveling time. In such a case, Pontryagin"s maximum principle is a powerful mathematical tool for solving this optimization problem. Three simulated results of removing a mobile object on a conveyor belt are presented; the object is grasped in motion by a planar three-link manipulator.