A Flexible Decision-Making Technology for Autonomous Robotics
The changes brought about by the ubiquity of drones and the forthcoming autonomous vehicles are just a foretaste of fundamental changes to come in the market of smart robots. Soon people will wear a variety of sensory-rich and active devices and work in smart environments (space habitats, home, offices, shopping malls, etc.) wired with sensors and actuators.
HybridLogic is a novel technology being developed by Menya Solutions Inc. to enable robotics systems to make sense of their environment, coordinate with other robots and users, anticipate events and plan and execute purposeful sequences of actions in order to timely handle novel situations for which they cannot be explicitly programmed.
Instead of opaque inference and decision making agents, HybridLogic Robotics agents provide explanations for their inferences to support effective human-robot collaborations both in understanding the situation and in planning actions.
Advanced Situation Understanding
HybridLogic combines sensor data, human input, and a priori knowledge to interpret the situation and infer the goals and plans of other agents in the environment. This information is used to support human-robot collaboration and to inform automated planning agents.
Advanced Automated Planning
HybridLogic enables robots to automatically synthesize or plan complex sequences of behaviours, combining manoeuvers and purposeful effectors such as manipulating and moving objects. Complex behaviours can be autonomously synthesized from lower-level behaviours conveyed by third-party robot operating systems. It uses adaptive coordination methods, enabling robots to automatically adjust their behaviours and information-exchange strategies to changing communication delays, bandwidth limitations, human interactions (e.g., collocated or remote), and other contextual parameters. HybridLogic supports a seamless integration of discrete and continuous planning (e.g., plan objects manipulations and robot motions simultaneously) as well as simultaneous macro-level (e.g., deciding and scheduling tasks) and micro-level planning (e.g., motion planning and onboard instrument manipulations). Each of these two hybrid features results in an agile and easy-of-use technology.