Conditional transition maps: Learning motion patterns in dynamic environments

Abstract

Abstract-In this paper we introduce a method for learning motion patterns in dynamic environments. Representations of dynamic environments have recently received an increasing amount of attention in the research community. Understanding dynamic environments is seen as one of the key challenges in order to enable autonomous navigation in real-world scenarios. However, representing the temporal dimension is a challenge yet to be solved. In this paper we introduce a spatial representation, which encapsulates the statistical dynamic behavior observed in the environment. The proposed Conditional Transition Map (CTMap) is a grid-based representation that associates a probability distribution for an object exiting the cell, given its entry direction. The transition parameters are learned from a temporal signal of occupancy on cells by using a localneighborhood cross-correlation method. In this paper, we introduce the CTMap, the learning approach and present a proof-ofconcept method for estimating future paths of dynamic objects, called Conditional Probability Propagation Tree (CPPTree). The evaluation is done using a real-world data-set collected at a busy roundabout.

Persistent localization and life-long mapping in changing environments using the Frequency Map Enhancement

“…In [16] the authors propose a new representation that models occupancy grid maps in the wavelet space in order to optimize the amount of information that has to be processed for path planning, and [17] presents a representation of the environment which models transitions of dynamic objects in the environment, by learning motion patterns from the temporal signal of occupancy in a cell.…”

SPENCER: A Socially Aware Service Robot for Passenger Guidance and Help in Busy Airports

“…For that, we have developed the Conditional Transition Map (CTMap, Kucner et al, 2013), a grid-based representation that models transitions of dynamic objects in the environment. For each cell x, CTMap learns the probability distribution of an object leaving to each neighboring cell, given the cell from which it entered into x.…”

Meta-rooms: Building and maintaining long term spatial models in a dynamic world

“…In the context of SLAM on 2D grid maps, the study of dynamics has lead to complex formulations of the traditional, static grid cell occupancy that take into account, in addition to occupancy, the dynamic element. For example, Saarinen et al use an independent Markov chain to model dynamics within grid cells [1]; Kuchner et al extend the traditional grid map approach with a Conditional Transition Map which models, for each grid cell, the transition probability of a dynamic object moving to one of its neighbouring cells, using a roundabout as a test scenario [2]. Walcott-Bryant et al address the problem of changing environments in a Pose-Graph SLAM system by augmenting a node in the graph with information concerning its dynamic state, and they maintain two separate mapsactive and dynamic -to better represent the environment [3].…”