Robot navigation in orchards with localization based on Particle filter and Kalman filter

Blok, Pieter M.,van Boheemen, Koen,van Evert, Frits K.,IJsselmuiden, Joris,Kim, Gook-Hwan Elsevier 2019 Computers and electronics in agriculture Vol.157 No.-

<P><B>Abstract</B></P> <P>Fruit production in orchards currently relies on high labor inputs. Concerns arising from the increasing labor cost and shortage of labor can be mitigated by the availability of an autonomous orchard robot. A core feature for every mobile orchard robot is autonomous navigation, which depends on sensor-based robot localization in the orchard environment. This research validated the applicability of two probabilistic localization algorithms that used a 2D LIDAR scanner for in-row robot navigation in orchards. The first localization algorithm was a Particle filter (PF) with a laser beam model, and the second was a Kalman filter (KF) with a line-detection algorithm. We evaluated the performance of the two algorithms when autonomously navigating a robot in a commercial Dutch apple orchard. Two experiments were executed to assess the navigation performance of the two algorithms under comparable conditions. The first experiment assessed the navigation accuracy, whereas the second experiment tested the algorithms’ robustness. In the first experiment, when the robot was driven with 0.25 m/s the root mean square error (RMSE) of the lateral deviation was 0.055 m with the PF algorithm and 0.087 m with the KF algorithm. At 0.50 m/s, the RMSE was 0.062 m with the PF algorithm and 0.091 m with the KF algorithm. In addition, with the PF the lateral deviations were equally distributed to both sides of the optimal navigation line, whereas with the KF the robot tended to navigate to the left of the optimal line. The second experiment tested the algorithms’ robustness to cope with missing trees in six different tree row patterns. The PF had a lower RMSE of the lateral deviation in five tree patterns. In three out of the six patterns, navigation with the KF led to lateral deviations that were biased to the left of the optimal line. The angular deviations of the PF and the KF were in the same range in both experiments. From the results, we conclude that a PF with laser beam model is to be preferred over a line-based KF for the in-row navigation of an autonomous orchard robot.</P> <P><B>Highlights</B></P> <P> <UL> <LI> Two localization algorithms compared in a Dutch apple orchard using Husky robot. </LI> <LI> Two experiments assessed navigation accuracy and navigation robustness. </LI> <LI> Particle filter outperformed Kalman filter on navigation accuracy and robustness. </LI> <LI> Algorithms are applicable for autonomous robot navigation using a 2D LIDAR scanner. </LI> </UL> </P>

A GPS/DR Data Fusion Method Based on the GPS Characteristics for Mobile Robot Navigation

Yuanliang Zhang,Kil To Chong 보안공학연구지원센터 2014 International Journal of Control and Automation Vol.7 No.10

In this paper we have considered the problem of outdoor mobile robot navigation using dead reckoning (DR) system and single GPS receiver. DR is a very simple and practical positioning technique. It is used in various positioning and navigation applications, especially for the mobile robot. DR can provide short term precise navigation information. But its errors will generally accumulate as the mobile robot continues to travel, and the calculated position of the mobile robot will become less and less accurate. For outdoor navigation application, GPS exhibit lots of advantages. It can provide real time and relatively accurate position data in spite of bad weather or other negative factors. But the big errors of civilian used single GPS receiver prevent it from applying to navigation for mobile robot alone. Differential GPS (DGPS) can be used to achieve an error of less than one meter but the costs are prohibitive in terms of commercializing it into the mass market. In this study, a cheap single GPS receiver cooperated with a DR system was used for the navigation system of a outdoors mobile robot in which a new GPS/DR data fusion method was utilized. This proposed fusion algorithm was based on the characteristics of the chosen single GPS receiver. The presented fusion algorithm does not bring much calculation burden and can provide accurate and robust navigation information for the mobile robot by adaptively switching between GPS/DR and DR when GPS lost the satellite signals. Simulation and experiment were performed to validate the effectiveness of the proposed fusion method and the good results showed its potential for outdoors mobile robot navigation.

Path Smoothing Extension for Various Robot Path Planners

Abhijeet Ravankar,Ankit A. Ravankar,Yukinori Kobayashi,Takanori Emaru 제어로봇시스템학회 2016 제어로봇시스템학회 국제학술대회 논문집 Vol.2016 No.10

Many path planning algorithms have previously been proposed for mobile robots to navigate from a start to a goal location in a given map. These planners generate a path which keeps a safe distance from the obstacles in the map. However, most of the global path planners generate a path with sharp and angular turns which is not desired for robot motion as robots must stop at these turns. A smooth path is desired for robot motion which allows the robot to move at nearly constant velocity. We present a novel path smoothing extension which uses the geometry of hypocycloids to smooth out the sharp and angular turns of the robot’s path and generates a smooth path for the robot to traverse. The proposed technique works as an ‘extension’ and can be used in conjunction with any of the previously proposed global path planners like D<SUP>*</SUP>, A<SUP>*</SUP>, Dijkstra, or PRM (Probabilistic Roadmap) planners. The proposed extension also generates ‘nodes’ on the robot’s path which can be used as points of retreat for the robot to avoid collision with other robots. Unlike other path smoothing algorithms which generates a wavy path for the robot and brings them close to the walls, the proposed path smoothing extension keeps straight paths of the robot straight, and smooths only the turns. We discuss the results in both simulated and real environments about the smooth paths generated by the proposed extension with different global path planners along with multirobot collision avoidance.

Supervised Control for Robot-Assisted Surgery Using Augmented Reality

Tzu-Hsuan Ho,Kai-Tai Song 제어로봇시스템학회 2020 제어로봇시스템학회 국제학술대회 논문집 Vol.2020 No.10

In this paper, we propose an AR-based robotic system that can plan and execute a trajectory based on a 3D medical model and allow the surgeon to supervise the execution of surgical process. In order to achieve image-guided surgery, a hand-eye calibration procedure is developed by using AprilTag to obtain the transformation between the workspace coordinate and the robot coordinate, and perform the image navigation task of the surgical robot to complete a drilling task. A procedure of robot supervised control is proposed to assign or modify the robot trajectory based on AR visualization and 3D model. A specified AR marker is used to project virtual objects in the AR glasses. We developed a robot registration algorithm to match the AR virtual coordinate system and the workspace coordinate system, and convert the planned trajectory into robot trajectory. Experimental results on the lab-built robotic system show that a user can adjust the position and orientation of the insertion point on a bone model, and transmit the trajectory information to the robot for execution. The proposed visualization-based robot navigation method has the potential to enhance the safety of surgical operation.

결합 가능한 복합 바퀴-다리 이동형 로봇에 관한 연구

이보훈(Bo-Hoon Lee),이창석(Chang-Seok Lee),김용태(Yong-Tae Kim) 한국지능시스템학회 2011 한국지능시스템학회논문지 Vol.21 No.6

로봇의 이동성 향상을 위해 다양한 환경에 적응할 수 있는 로봇의 연구 개발이 활발하게 진행되고 있다. 본 논문에서는 휠(wheel)과 다리(Leg)기반 변형이 가능하고, 로봇간 상호 결합이 가능한 복합 이동형 로봇을 제안하였다. 복합 이동형 로봇은 로봇간 결합을 위해 페그 모듈과 컵 모듈을 로봇의 전면과 후면에 각각 장착하고, 주행과 보행이 가능하도록 구현하였다. 다양한 지형에서 이동성을 향상을 위해 임베디드 영상기반 결합 및 분리 알고리즘을 제안하였으며, 로봇간 결합을 통해 끊어진 도로와 비평탄 지형에서의 결합 이동 방법을 제안하였다. 제안한 방법은 로봇의 전면과 밑면에 장착된 PSD 센서를 이용하여 지형을 인식하고, 지형에 맞은 극복 알고리즘을 통해 로봇간 협력을 통해 이동성을 향상시킨다. 제안한 방법들은 임베디드시스템 기반의 복합 주행 이동형 로봇을 실제 제작하여 실험 통해 성능을 검증하였다. There are many researches to develop robots that improve its mobility to adapt in various uneven environments. In the paper, a hybrid mobile robot that can dock with the other robot and transforms between wheeled robot and legged robot is proposed. The hybrid mobile robot platform has docking device with a peg and a cup module. In addition, the robot is possible to walk and drive according to condition of the road. A navigation algorithm of the hybrid mobile robot is proposed to improve the mobility of robots using docking algorithm based on image processing on the broken road and uneven terrain. The proposed method recognizes road condition through PSD sensor attached in front and bottom of the robot and selects an appropriate navigation method according to terrain surface. The proposed docking and navigation methods are verified through experiments using hybrid mobile robots.

이동 로봇의 물체 인식과 주행을 위한 로봇 지식 체계

임기현(Gi Hyun Lim),서일홍(Il Hong Suh) 대한전자공학회 2007 電子工學會論文誌-CI (Computer and Information) Vol.44 No.6

본 논문에서는 이동 로봇의 다계층으로 로봇 지식 체계를 구축함으로써 실생활 환경에서 잡음이 섞인 센서 때문에 소실되거나 잃어버리거나 가려진 정보를 찾아낼 수 있는 추론(inference)할 수 있는 로봇 지식을 구현하고자 한다. 로봇 지식 체계는 4개의 지식 계층과 2종류의 규칙 (rule) 과 공리 (Axiom)으로 구성되어 있다. 인지, 모델, 정황, 활동의 4 개의 지식 계층(KClass) 으로 구성된다. 각각의 지식 계층은 3개의 지식 층 (KLevel) 과 3개의 온톨로지 층 (OLayer) 으로 구성된다. 3개의 지식층은 하위 층, 중간, 상위 지식층이고, 3 개의 온톨로지 층은 메타 온톨로지, 온톨로지 스키마, 온톨로지 인스턴스 층이다. 공리는 각 온톨로지 층 내에서 온톨로지 요소인 개념간의 관계를 표현하고, 2종류의 규칙은 서로 다른 온톨로지 층간, 서로 다른 지식 계층 간의 연관을 각각 표현한다. 따라서 이러한 특징의 로봇의 하위 수준의 센서 정보에서 상위 수준의 의미 정보를 통합 할 수 있도록 하고, 통합된 지식을 가지고 이웃한 층간의 단방향 추론 및 몇 개의 층들 간의 양방향 추론을 통해 불확실하고 부분적인 정보에 대한 질문에 응답할 수 있다. 이러한 우리의 로봇 지식 체계의 유용성이 물체 인식과 주행을 위한 여러 실험을 통하여 검증할 수 있다. This paper introduces a robot knowledge framework which is represented with multiple classes, levels and layers to implement robot intelligence at real environment for mobile robot. Our root knowledge framework consists of four classes of knowledge (KClass), axioms, rules, a hierarchy of three knowledge levels (KLevel) and three ontology layers (OLayer). Four KClasses including perception, model, activity and context class. One type of rules are used in a way of unidirectional reasoning. And, the other types of rules are used in a way of bi-directional reasoning. The robot knowledge framework enable a robot to integrate robot knowledge from levels of its own sensor data and primitive behaviors to levels of symbolic data and contextual information regardless of class of knowledge. With the integrated knowledge, a robot can have any queries not only through unidirectional reasoning between two adjacent layers but also through bidirectional reasoning among several layers even with uncertain and partial information. To verify our robot knowledge framework, several experiments are successfully performed for object recognition and navigation.

Caterpillar Equipped Mobile Robot Navigation with Dynamic Windows Approach for Outdoor Agriculture Application

( Tianyuan Guan ),( Chen Tean ),( Sangeon Oh ),( Kyeonghwan Lee ) 한국농업기계학회 2019 한국농업기계학회 학술발표논문집 Vol.24 No.2

Autonomous robot have great potential to deal with various kinds of fieldwork. In spite of wheeled robot can adapt most landforms, it has to face to stability and trafficability problems in tough terrains. In order to apply to complicate outdoor environments, we develop a caterpillar equipped robot system with simplified Dynamic Window Approach for agriculture application. The caterpillar equipped robot system composed by a localization system which is integration of RTK-GPS, IMU and IMU auto-calibrate device, a Robot Operating System(ROS) high-level controller and a base controller. The caterpillar mobile robot has features of high traction and high mobility. Since it has differential drive architecture, different kinds of navigation algorithm can be easily applied to it. Dynamic Windows Approach(DWA) is a local navigation algorithm, which can select optimized velocity for robot by estimating its current heading and distance to goal, and it also can generate smooth path for robot. As the results of experiments, our robot can cruise certain path accurately even in rough terrain, and it can also correct orientation by itself periodically, it can satisfy the demand of outdoor agriculture usage.

Integrated Path Planning and Collision Avoidance for an Omni-directional Mobile Robot

Dong Hun Kim 한국지능시스템학회 2010 INTERNATIONAL JOURNAL of FUZZY LOGIC and INTELLIGE Vol.10 No.3

This paper presents integrated path planning and collision avoidance for an omni-directional mobile robot. In this scheme, the autonomous mobile robot finds the shortest path by the descendent gradient of a navigation function to reach a goal. In doing so, the robot based on the proposed approach attempts to overcome some of the typical problems that may pose to the conventional robot navigation. In particular, this paper presents a set of analysis for an omni-directional mobile robot to avoid trapped situations for two representative scenarios: 1) Ushaped deep narrow obstacle and 2) narrow passage problem between two obstacles. The proposed navigation scheme eliminates the nonfeasible area for the two cases by the help of the descendent gradient of the navigation function and the characteristics of an omni-directional mobile robot. The simulation results show that the proposed navigation scheme can effectively construct a path-planning system in the capability of reaching a goal and avoiding obstacles despite possible trapped situations under uncertain world knowledge.

다중 센서 기반 무한궤도형 이동로봇의 실외 자율 주행시스템

방효석,김용태,이창준,박민홍,조재훈 한국지능시스템학회 2022 한국지능시스템학회논문지 Vol.32 No.2

This paper proposes an outdoor navigation system of a caterpillar mobile robot based on the multiple sensors. The real-time state estimation and the obstacle avoidance algorithm is applied for the proposed navigation system using a 3D LIDAR sensor. A caterpillar mobile robot is designed to enable stable driving in the uneven terrain with many small obstacles. In order to improve the performance of state estimation of the proposed navigation system, the driving accuracy and localization of the robot is improved by using encoder and IMU data, the converted two-dimensional coordinate data of RTK GPS. An individual module is designed to verify the driving performance of the autonomous driving system to the destination and an outdoor experiment was conducted by developing a prototype robot. The experiment results show that the caterpillar mobile robot autonomously navigate in the uneven environment. 본 논문에서는 무한궤도형 이동로봇의 다중 센서 기반 실외 자율주행 시스템을 제안한다. 제안된 자율주행시스템은 로봇의 실시간 위치 추정이 가능하며, 3차원 라이다 센서를 통해장애물 감지 및 회피 주행알고리즘을 적용하였다. 소규모 장애물이 많은 비평탄 지형에서안정적인 주행이 가능하도록 무한궤도를 사용한 로봇을 설계하였다. 주행시스템의 위치 추정 성능을 개선하기 위해 엔코더와 IMU 데이터에 RTK GPS의 변환된 2차원 좌표 데이터를 융합하여 로봇의 위치 인식 및 주행 정밀도를 높였다. 자율주행시스템의 목적지까지 주행 성능을 검증하기 위해 개별 모듈을 설계하였고, 시제품을 제작하여 실외 실험을 수행하였다. 실험 결과에서 비평탄 지형에서 무한궤도형 이동 로봇의 자율주행이 가능함을 확인하였다

Corridor Navigation of the Mobile Robot Using Image Based Control

Han, Kyu-Bum,Kim, Hae-Young,Baek, Yoon-Su The Korean Society of Mechanical Engineers 2001 JOURNAL OF MECHANICAL SCIENCE AND TECHNOLOGY Vol.15 No.8

In this paper, the wall following navigation algorithm of the mobile robot using a mono vision system is described. The key points of the mobile robot navigation system are effective acquisition of the environmental information and fast recognition of the robot position. Also, from this information, the mobile robot should be appropriately controlled to follow a desired path. For the recognition of the relative position and orientation of the robot to the wall, the features of the corridor structure are extracted using the mono vision system, then the relative position, the offset distance and steering angle of the robot from the wall, is derived for a simple corridor geometry. For the alleviation of the computation burden of the image processing, the Kalman filter is used to reduce search region in the image space for line detection. Next, the robot is controlled by this information to follow the desired path. The wall following control scheme by the PD control scheme is composed of two control parts, the approaching control and the orientation control, and each control is performed by steering and forward-driving motion of the robot. To verify the effectiveness of the proposed algorithm, the real time navigation experiments are performed. Through the result of the experiments, the effectiveness and flexibility of the suggested algorithm are verified in comparison with a pure encoder-guided mobile robot navigation system.