Intel Corporation

Venkataraman Natarajan

Gagan Acharya

Jaroslaw Sydir

Amit S Baxi

Kumar Ranganathan

Apoorv Vyas

Shagaya Mageshkumar Vincent

Prasad Ghatigar

Ramya M

Arjun K G

Santosh Shukla

Nishtha Namdeo Vaidya

The present disclosure provides describes to train a multi policy ML model to control robots in a multi-robot system in collaborating to perform a task. For example, trajectories associated with manipulating an object to perform the collaborative task can be determined and an ML model trained to output control actions for the robots in the multi-robot system to collaborate to complete the task.

Training collaborative robots through user demonstrations

Methods and apparatus for complex assembly via autonomous robots using reinforcement learning action primitives are disclosed. An example apparatus includes a construction manager and a movement manager. The construction manager is to determine sequences of reinforcement learning (RL) action primitives based on object location goals and associated assembly goals determined for respective ones of objects depicted in an imaged assembly of objects. The movement manager is to command a robot to construct a physical assembly of objects based on the sequences of RL action primitives. The physical assembly of objects is to correspond to the imaged assembly of objects.

Methods and apparatus for complex assembly via autonomous robots using reinforcement learning action primitives

Techniques disclosed for accurately predicting the occurrence of anomalous sensor readings within a sensor network and advantageously using these predictions to limit the amount of power used by relay nodes within the sensor network. Some examples analyze spatial and temporal characteristics of anomalous sensor readings to predict future occurrences. In these examples, the relay nodes operate in a reduced power mode for periods of time in which anomalous sensor readings are not predicted to occur. Also, in these examples, only relay nodes in a path between a sensor predicting an anomalous reading and a gateway of the sensor network operate in full power mode. This feature allows other relay nodes to remain in the reduced power mode even when an anomalous sensor reading is predicted elsewhere in the sensor network.

Scheduling of network relays

According to an aspect of the present invention, a quotient of a dividend divided by a divisor may be determined after reducing the dividend, divisor, and the remainder by using operations such as add, subtract, multiply, shift, AND which may result in reduced processor cycles (time).

Division algorithm

Growing community of inventors

Bengaluru, India

Venkataraman Natarajan