Gambit: An autonomous chess-playing robotic system

Abstract

This paper presents Gambit, a custom, mid-cost 6-DoF robot manipulator system that can play physical board games against human opponents in non-idealized environments. Historically, unconstrained robotic manipulation in board games has often proven to be more challenging than the underlying game reasoning, making it an ideal testbed for small-scale manipulation. The Gambit system includes a low-cost Kinectstyle visual sensor, a custom manipulator, and state-of-the-art learning algorithms for automatic detection and recognition of the board and objects on it. As a use-case, we describe playing chess quickly and accurately with arbitrary, uninstrumented boards and pieces, demonstrating that Gambit's engineering and design represent a new state-of-the-art in fast, robust tabletop manipulation.

Design and Implementation of a Computer Vision System for an Autonomous Chess-Playing Robot

“…This issue results in specific designs that do not incorporate, in the design phase, the complexities introduced by the use of arbitrary pieces and boards on real environments. The implementations in Gambit [6] and Baxter [7] seems to be the closest systems to the one here presented in terms of functionality. Gambit's use SVMs to learn the piece types and besides it uses two cameras including depth information.…”

“…Some applications, such as in [5], use a simplified XY linear slide system. However, those autonomous systems commonly use instrumented boards and parts specially codesigned with the manipulator in order to simplify the tasks of piece detection and manipulation [6]. This issue results in specific designs that do not incorporate, in the design phase, the complexities introduced by the use of arbitrary pieces and boards on real environments.…”

Accelerating imitation learning through crowdsourcing

“…1) designed at Intel Labs Seattle. Gambit is well-suited to smallobject tabletop manipulation tasks and has previously been shown to perform well in tasks with humans in the loop [16].…”

“…Gambit can pick and place blocks at pre-specified locations 2 . Even though Gambit is precise enough to play a game of chess [16], building 2D object models turns out to be a nontrivial task because humans typically build models in which blocks abut each other. This close block-to-block proximity means that for the robot, the difficulty of the task increases with the number of blocks in the model, e.g.…”

Extending the reach of social touch through haptic technology

“…Indeed, it is incredibly difficult to design a robot that has a sense of touch akin to that of a human, let alone a robot with the adaptability with which a human can operate his or her extremities to move the small, smooth pieces on a Go board [429,435]. Perhaps that is something Go players can take pride in, then, especially since the manipulation of chess pieces has already been mastered by robots [346].…”