Iterative matrix processor based implementation of real-time model predictive control
First Claim
Patent Images
1. A system comprising:
- a plant controlled by real-time model predictive control;
an iterative matrix processor configured to perform computations on vectors and matrices stored in a logarithmic format;
a general purpose processor in communication with the plant and the iterative matrix processor to transfer information between the plant and the iterative matrix processor, the general purpose processor being configured to issue a sequence of instructions to the iterative matrix processor to implement the model predictive control; and
wherein each instruction of the sequence of instructions initiates a series of element-by-element operations in the logarithmic format on scalars, vectors and matrices stored in the iterative matrix processor.
1 Assignment
0 Petitions
Accused Products
Abstract
A system for embedding real-time Model Predictive Control (MPC) in a System-on-a-Chip (SoC) devices is provided. In the system, a microprocessor is connected to an auxiliary unit or application-specific matrix coprocessor. The microprocessor can control the operation of the MPC algorithm, i.e., carry out the tasks of input/output for the MPC algorithm, initialize and send the appropriate commands to auxiliary unit and receive back the optimal control moves or instructions from auxiliary unit. The auxiliary unit can operate as a matrix coprocessor by executing matrix operations, e.g. addition, multiplication, inversion, etc., required by the MPC algorithm.
117 Citations
18 Claims
-
1. A system comprising:
-
a plant controlled by real-time model predictive control; an iterative matrix processor configured to perform computations on vectors and matrices stored in a logarithmic format; a general purpose processor in communication with the plant and the iterative matrix processor to transfer information between the plant and the iterative matrix processor, the general purpose processor being configured to issue a sequence of instructions to the iterative matrix processor to implement the model predictive control; and wherein each instruction of the sequence of instructions initiates a series of element-by-element operations in the logarithmic format on scalars, vectors and matrices stored in the iterative matrix processor. - View Dependent Claims (2, 3, 4, 5)
-
-
6. A controller to implement a real-time model predictive control algorithm, the controller comprising:
-
a microprocessor to execute a portion of the model predictive control algorithm; and an iterative matrix processor configured to receive an instruction from the microprocessor to perform at least one matrix calculation of the model predictive control algorithm, and the iterative matrix processor being configured to operate using a logarithmic number system. - View Dependent Claims (7, 8, 9, 10)
-
-
11. A model predictive controller comprising:
-
a general purpose processor to execute a model predictive control algorithm; an iterative matrix processor in communication with the general purpose processor, the iterative matrix processor being configured to execute at least one matrix calculation required by the model predictive control algorithm in an iterative process in response to receiving an instruction from the general purpose processor; and wherein the iterative matrix processor performs the at least one matrix calculation using a logarithmic number system. - View Dependent Claims (12, 13, 14, 15, 16, 17, 18)
-
Specification