China
Africa
Explore chapters and articles related to this topic
Workspace Sharing Assembly Robots: Applying IEC 61499 to System Integration and Application Development
Published in Alois Zoitl, Thomas Strasser, Distributed Control Applications, 2017
Matthias Plasch, Ebenhofer Gerhard, Michael Hofmann, Martijn Rooker, Sharath Chandra Akkaladevi, Andreas Pichler
The results of the vision module are used to plan and calculate collision-free robot manipulation paths to enable handling of the detected objects. Based on predefined grasp as well as deposit points on the CAD model of the objects, the manipulation planner determines how the object can be grasped. The path planning algorithm is based on the rapidly-exploring random trees (RRT) approach [18][6]. In order to enable collision-free cooperation of both robots, a virtual wall to separate the workspace into two areas is dynamically inserted into the working area. Using the limited workspace, the first robot tries to find a collision-free path for grasping and manipulating objects. Among both robots, one is considered the master and is allowed to shift the wall until it finds a valid collision-free path. The other robot (slave) may use the remaining workspace for path planning.
Warehouse Automation: An Example
Published in Laxmidhar Behera, Swagat Kumar, Prem Kumar Patchaikani, Ranjith Ravindranathan Nair, Samrat Dutta, Intelligent Control of Robotic Systems, 2020
Laxmidhar Behera, Swagat Kumar, Prem Kumar Patchaikani, Ranjith Ravindranathan Nair, Samrat Dutta
In the case of industrial manipulators where one does not have access to internal motor controllers, motion planning refers to providing suitable joint angle position (or velocity) trajectories needed for taking the robot from one pose to another. In other words, motion planning becomes a path planning problem which is about finding a way to point poses between the current pose and the desired end-effector pose. The problem of generating collision free paths for manipulators with increasingly larger number of links is quite complex and has attracted considerable interest over last couple of decades. Readers can refer to [351] for an overview of these methods. These methods could be primarily divided into two categories - local and global. Local methods start from a given initial configuration and step toward final configuration by using local information of the workspace. Artificial potential field-based methods [352] [353] [354] are one such category of methods where the search is guided along the negative gradient of artificially created vector fields. On the other hand, global methods use search algorithms over the entire workspace to find suitable paths. Some of the examples of global methods are probabilistic roadmaps (PRM) [355] [356]and cell-decomposition based C-Space methods [357] [358]. Rapidly exploring random tree (RRT) [359] is one of the most popular PRM method used for path planning. Many of these state-of-the-art algorithms are available in the form of the open motion planning library (OMPL) [289] which has been integrated into several easy-to-use software packages like Moveit! [288], Kautham [360], and OpenRave [361].
Conceptual Design of the Pedestrian Bridge
Published in Structural Engineering International, 2022
Pengzhen Lu, Yutao Zhou, Qun Lu, Jiahao Wang, Qingtian Shi, Dengguo Li
When a robot system is adopted in the assembly process, the robot receives real-time information about the bridge construction project transmitted by the intelligent system and completes the automatic grasping of parts, positioning and installation by inputting the part model and position coordinates to be assembled, and then enters the next cycle. For example, a Rapidly-exploring Random Trees (RRT) path planning algorithm can be adopted to realize the obstacle avoidance application of a robot. Target recognition of robots can be based on the Kinect® sensor algorithm.
Methodology and decentralised control of modularised changeable conveyor logistics system
Published in International Journal of Computer Integrated Manufacturing, 2019
Boqian Yu, Nan Xie, Beirong Zheng, Deji Chen
The Rapidly-exploring Random Tree (RRT) proposed in (LaValle 1998) is a randomised structure widely used for path planning problems. The basic idea of RRT is to sample a point randomly in the entire configuration space and to expand towards it from the nearest explored point. To solve the problem that the exploration expands too averagely for the entire configuration space, RRT-GoalBias was proposed in (LaValle & Kuffner, Rapidly-exploring random trees: Progress and prospects, 2001), adding inclination for the target location.
Coarse grid partition to speed up A* robot navigation
Published in Journal of the Chinese Institute of Engineers, 2020
Chien-Yen Wang, Chan-Yun Yang, Shadi Banitaan, Chaomin Luo, Sainzaya Galsanbadam
There are a number of algorithms that already exist for producing the optimal traversals and given transition costs. A* algorithm (Stentz 1995; Maxim, Geoff, and Sebastian 2003; Koenig and Likhachev 2005) plans an optimal path based on grid-based map representation, and moves the robot along the planned path until it reaches the goal. A distance transformation using prior map information is adopted to represent the relevance. Many algorithmic principles have been proposed for resolving the path planning problem. For example, the potential field approach (Lai et al. 2007; Kala and Warwick 2013; Lin and Chuang 2010), which was inspired by the concept of electrical charges, is able to handle the problem well, but the heuristic in the approach would spend a lot of time in an explored solution. A probabilistic-based approach, namely the Probabilistic Road Map (PRM) (Kavraki et al. 1996), generates quantitatively random samples from the configuration space of the robot for exploring a feasible path from a starting configuration S to a goal configuration G by querying the dense point and edge configurations while avoiding collisions. Having been proven to be probabilistically complete, the PRM solutions are not uniform in gaining the path plans in every run due to their probabilistic orientation. Similar to the workflow of PRM, Rapidly-exploring Random Tree (RRT), which operates alternatively by growing a tree in the configuration space instead of the probabilistic random points and edges, largely prunes away improper or unnecessary entities and uses less time (LaValle and Kuffner 2000). Though there are benefits from the tree because the RRT and its enhanced work of bidirectional RRT (Jordan and Perez 2013; Xue et al. 2017) ensure the optimal tree by double checking both forward searching from S and backward from G, one disadvantage is the asymptotic convergence greatly depends on the number of examples. In the approach using a combination of the A* algorithm and Fuzzy Inference, the A* algorithm does the higher level planning by working on a less detailed map. The lower level planning is done by the Fuzzy Inference System (Kala, Shukla, and Tiwari 2010; Dirik, Kocamaz, and Dönmez 2017). Nevertheless, this approach, based on the properties of probability, also cannot obtain a unique and optimal solution. In addition, neural networking (Singha, Ray, and Samaddar 2017) and evolutionary programming (Contreras-Cruz, Ayala-Ramirez, and Hernandez-Belmonte 2015; Huang, Wu, and Huang 2014) are able to solve this problem. However, those algorithms have more iterations, which will increase the execution time.