No abstract

Four major approaches to robot motion planning in dynamic environments are discussed: probabilistic robot, probabilistic collision state (PCS), partially closed-loop receding horizon control (PCLRHC) and gross hidden Markov model (GHMM). A comparison of three mapping techniques, Kalman filter, expectation and maximization algorithm and Markov model, is presented. The PCS method is the probabilistic extension of inevitable collision state, which is found to be the safest motion planning method. The concept of open-loop and partially closed-loop receding horizon control (OLRHC and PCLRHC) is compared critically, and the algorithms are benchmarked. GHMM is the best suited method for environments with limited space and dynamic environment due to human interactions. GHMM parameters and structure are evaluated using an incremental "learn-and-predict" approach. For exploring GHMM, we simulated a cafeteria with eight tables to be served by a robot, considering three different arrangements of tables along with convex and concave obstacles, and obtained the path length and time taken for a Hamiltonian path. During the simulation, it was observed that for a given static or dynamic environment, the concavity of the obstacles is what makes the scenario a complex one.

A collision free path to a target location in a random farm is computed by employing a probabilistic roadmap (PRM) that can handle static and dynamic obstacles. The location of ripened mushrooms is an input obtained by image processing. A mushroom harvesting robot is discussed that employs inverse kinematics (IK) at the target location to compute the state of a robotic hand for holding a ripened mushroom and plucking it. Kinematic model of a two-finger dexterous hand with 3 degrees of freedom for plucking mushrooms was developed using the Denavit-Hartenberg method. Unlike previous research in mushroom harvesting, mushrooms are not planted in a grid or some pattern, but are randomly distributed. No human intervention is required at any stage of harvesting.

The fundamental problem of robot motion planning in a dynamic environment (RMPDE) is to find an optimal collision-free path from the start to the goal in a dynamic environment. Our literature survey of over 100 papers from the last four decades reveals that there are more than 30 models of RMPDE, and there is no benchmarking criterion to select one that is the best in a given situation. In this context, generating a regression-based model with 10 attributes is the first and foremost contribution of our research. Given a highly human-interactive environment like a cafeteria or a bus stand, the gross hidden Markov model has special importance for modeling a robot path. A variant of the growing hidden Markov model for a serving robot in a cafeteria is the second contribution of this paper. We simulated the behavior of GHMM in a cafeteria with static and dynamic obstacles (static obstacles were both convex and concave) and with three different arrangements of the tables and obstacles. Robots have been employed in mushroom harvesting. A novel proposition discussed in this paper is probabilistic road map planning for a robot that finds an optimum path for reaching the ripened mushrooms in a randomly planted mushroom farm and a dexterous hand to pluck the selected mushrooms by employing inverse kinematics. Further, two biologically inspired meta-heuristic algorithms, ant colony optimization, and firefly has been studied for their application to latex collection. The simulation results with this environment show that the firefly algorithm outperforms ant colony optimization in the general case. Finally, we have proposed a few pointers for future research in this domain.  The compilation and comparison of various approaches to robot motion planning in highly dynamic environments, and the simulation of a few models for some typical scenarios, have been the contributions of this paper.

collision-free path to a destination position in a random farm is determined using a probabilistic roadmap thatcan manage static and dynamic obstacles. The position of ripening mushrooms is a result of picture processing.A mushroom harvesting robot is explored that uses inverse kinematics at the target position to compute the stateof a robotic hand for grasping a ripening mushroom and plucking it. The Denavit–Hartenberg approach was usedto create a kinematic model of a two-finger dexterous hand with three degrees of freedom for mushroom picking.Unlike prior experiments in mushroom harvesting, mushrooms are not planted in a grid or design but are randomlyscattered. At any point throughout the harvesting process, no human interaction is necessary