Case-based reactive navigation: a method for on-line selection and adaptation of reactive robotic control parameters

Abstract

We present a new line of research investigating on-line adaptive reactive control mechanisms for autonomous intelligent agents. We discuss a case-based method for dynamic selection and modification of behavior assemblages for a navigational system. The case-based reasoning module is designed as an addition to a traditional reactive control system, and provides more flexible performance in novel environments without extensive high level reasoning that would otherwise slow the system down. The method is implemented in the ACBARR (case-based reactive robotic) system and evaluated through empirical simulation of the system on several different environments, including "box canyon" environments known to be problematic for reactive control systems in general.

Navigation Among Humans

“…CBR allows recalling and interpreting past experiences, as well as generating new cases to represent knowledge from new experiences. CBR has previously proven successful in solving spatio-temporal problems in robotics in Jurisica & Glasgow (1995); Likhachev & Arkin (2001); Ram et al (1997), but not to estimate intentions of humans. Other methods, like Hidden Markov Models, have been used to learn to identify the behaviour of humans Kelley et al (2008).…”

Assistive technologies for the new generation of senior citizens: the SHARE-it approach

“…A CBR system cycle to solve a new problem consists of four steps (Aamodt and Plaza, 1994 CBR has been used in mobility platform navigation before, but typically for high level planning rather than to accomplish reactive behaviours (Branting and Aha, 1995;Likhachev et al, 2002;Ram et al, 1993;Urdiales et al, 2006). In our case, each time a new input instance is detected, the CBR captures a new case, coupling this instance with the user's joystick output and using the local efficiencies (directiveness, smoothness and safety) to rank how good the case is.…”

Behavior-based Robot Control Using Fuzzy Discrete Event System

“…However, a common problem arises in command fusion techniques, when competing behaviors issue conflicting control commands resulting in local minima or stagnant situations of the robot [9,46]. This problem is addressed by weighted decision-making techniques [47,48,49,50,51,52,53,54,55], where different sets of rules are used to weight the conflicting commands according to environmental contexts, thus, reducing the possibility of producing local minima.…”

“…(b) A novel behavior-based approach of mobile robot navigation is proposed using FL-based coordination technique for motor schema [41] based behaviors. This method is presented as an alternative of the case-based navigation using motor schema [47]. It shows better performance in eliminating the shortcomings of the conventional motor schema based navigation.…”

“…A case-based approach was also proposed by Ram et al [47], where the environment is divided into a set of special cases and the force vectors were weighed according to a particular case to overcome the trap situations. However, the number of cases increases indefinitely with the environment complexity.…”