View planning for exploration via maximal C-space entropy reduction for robot mounted range sensors

We introduced the concept of C-space entropy recently as a measure of knowledge of configuration space (C-space) for sensor-based exploration and path planning for general robot–sensor systems. The robot plans the next sensing action to maximally reduce the expected C-space entropy, also called the Maximal expected Entropy Reduction (MER) criterion. The resulting view planning algorithms showed significant improvement of exploration rate over physical space-based criteria. However, this expected C-space entropy computation made two idealized assumptions: (i) that the sensor field of view (FOV) is a point and (ii) that there are no occlusion (or visibility) constraints, i.e., as if the sensor can sense through the obstacles. We extend the expected C-space entropy formulation where these two assumptions are relaxed, and consider a range sensor with non-zero volume FOV and occlusion constraints, thereby modeling a realistic range sensor. Planar simulations and experimental results on the SFU Eye-in-Hand system show that the new formulation results in further improvement in C-space exploration efficiency over the point FOV sensor-based MER formulation.     

A Framework for Automatic Generation of a Contact State Graph for Robotic Assembly

Dynamic manipulation of an active object is introduced as a general model of hopping and juggling tasks. In this setting, juggling and hopping are two extreme cases of this general model. Behavioral resemblance of these two tasks is afterwards extended to a detailed mathematical analogy between them. Then the analogy is exploited to develop a unified and abstract planning framework for juggling and hopping. To this end, dynamic manipulation of an active object is decomposed into three distinct phases and two transitions: Carry I, Free flight and Carry II phases. These phases are analogous to Lift off, Free flight and Touch down in hopping. In the next step, a mathematical model for each phase is developed. It is shown that dynamic grasp (in Carry phases of juggling) and foot stability (in Support phases of hopping) conditions share similar sets of dynamic equations. Accordingly, Lift off/Release and Touch down/Catch conditions in hopping/juggling are derived. It is shown that analogous strategies can be developed for Lift off and Release. The analogy is held for Touch down and Catch conditions as well. It is discussed that in the planning framework the initial and the goal configurations of the three phases are set in a model-based and forward manner. To do so, Touch down/Landing time, Free flight duration and robot/object maneuvers during Free flight are used as free parameters for planning in order to ensure foot stability in hopping and dynamic grasp in juggling along with other constraints.     

Three-Dimensional Path Planning of a Climbing Robot Using Mixed Integer Linear Programming

The City-Climber robot is a novel wall-climbing robot developed at The City College of New York that has the capability to move on floors, climb walls, walk on ceilings and transit between them. In this paper, we first develop the dynamic model of the City-Climber robot when it travel on different surfaces, i.e., floors, walls and ceilings, respectively. Then, we present a path planning method for the City-Climber robot using mixed integer linear programming (MILP) in three-dimensional (3-D) building environments that consist of objects with primitive geometrical shapes. MILP provides an optimization framework that can directly incorporate dynamic constraints with logical constraints such as obstacle avoidance and waypoint selection. In order to use MILP to solve the obstacle avoidance problem, we simplify and decouple the robot dynamic model into a linear system by introducing a restricting admissible controller. The decoupled model and obstacle can be rewritten as a linear program with mixed-integer linear constraints that account for the collision avoidance. A key benefit of this approach is that the path optimization can be readily solved using the AMPL and CPLEX optimization software with a MATLAB interface. Simulation results show that the framework of MILP is well suited for path planning and obstacle avoidance problems for the wall-climbing robot in 3-D environments.     

Generic Motion Planner for Robot Multi-fingered Manipulation

This paper addresses the dexterous manipulation planning problem, which deals with motion planning for a multi-fingered hand manipulating objects among static obstacles, under quasi-static movement assumption. We propose a general manipulation approach able to compute object and finger trajectories, as well as the finger relocation sequence, in order to link any two given configurations of the composite system hand + object. It relies on a topological property that characterizes the existence of solutions in the subspace of configurations where the object is grasped by the n fingers. This property helps reduce the problem by structuring the search space. The developed planner captures in a probabilistic roadmap the connectivity of submanifolds of the composite configuration space. The answer to the manipulation planning query is then given by searching a path in the computed graph. Simulation experiments are reported for different multi-fingered manipulation task examples showing the efficiency of the proposed method.     

Triple RRTs: An Effective Method for Path Planning in Narrow Passages

Although Rapidly-exploring Random Trees (RRTs) have been successfully applied in path planning of robots with many degrees of freedom under non-holonomic and differential constraints, rapidly identifying and passing through narrow passages in a robot's configuration space remains a challenge for RRTs-based planners. This paper presents a novel two-stage approach to address the problem of multi-d.o.f. robot path planning in high-dimensional configuration space with narrow corridors. The first stage introduces an efficient sampling algorithm called Bridge Test to find a global roadmap that identifies the critical region. The second stage presents two varieties of RRTs, called Triple-RRTs, to search for a local connection under the guidance of the global landmark. The two-stage strategy keeps a fine balance between global heuristics and local connection, resulting in high performance over the previous RRTs-based path planning methods. We have implemented the Triple-RRTs planners for both rigid and articulated robots in two- and three-dimensional environments. Experimental results demonstrate the effectiveness of the proposed method.     

Local path planning using a new artificial potential function composition and its analytical design guidelines

This paper presents a new framework for path planning based on artificial potential functions (APFs). In this scheme, the APFs for path planning have a multiplicative and additive composition between APFs for goal destination and APFs for obstacle avoidance, unlike conventional composition where the APF for obstacle avoidance is added to the APF for goal destination. In particular, this paper presents a set of analytical guidelines for designing potential functions to avoid local minima for a number of representative scenarios based on the proposed framework for path planning. Specifically the following cases are addressed: (i) a non-reachable goal problem (a case in which the potential of the goal is overwhelmed by the potential of an obstacle), (ii) an obstacle collision problem (a case in which the potential of the obstacle is overwhelmed by the potential of the goal) and (iii) a narrow passage problem (a case in which the potential of the goal is overwhelmed by the potential of two obstacles). The example results for each case show that the proposed scheme can effectively construct a path-planning system with the capability of reaching a goal and avoiding obstacles despite possible local minima.     

New repulsive potential functions with angle distributions for local path planning

This paper presents new repulsive potential functions (RPFs) for point robot path planning. In this scheme, the RPF for path planning has a different magnitude at each direction of a RPF based on the angle between a goal and an obstacle, unlike a conventional RPF in which the same magnitude at each direction is obtained. In doing so, the RPF attempts to overcome some of the typical problems that may arise with the conventional RPF. In particular, this paper presents a set of analyses for designing potential functions to avoid local minima for a number of representative scenarios. Specifically, the following cases are addressed: (i) a non-reachable goal problem (a case in which the potential of the goal is overwhelmed by the potential of an obstacle), (ii) an obstacle collision problem (a case in which the potential of the obstacle is overwhelmed by the potential of the goal) and (iii) a narrow passage problem (a case in which the potential of the goal is overwhelmed by the potential of two obstacles). The proposed RPF scheme eliminates the non-feasible area for the three cases by the help of an angle-varying magnitude between a goal and an obstacle. The example results show that the proposed RPF scheme can effectively construct a path-planning system with the capability of reaching a goal and avoiding obstacles despite possible local minima.     

The meaning of action: a review on action recognition and mapping

In this paper, we analyze the different approaches taken to date within the computer vision, robotics and artificial intelligence communities for the representation, recognition, synthesis and understanding of action. We deal with action at different levels of complexity and provide the reader with the necessary related literature references. We put the literature references further into context and outline a possible interpretation of action by taking into account the different aspects of action recognition, action synthesis and task-level planning.