Undergraduate Teaching 2018-19

Not logged in. More information may be available... Login via Raven / direct.

Engineering Tripos Part IIA Project, GB3: RISC-V Processor, 2018-19

Leader

Dr P Stanley-Marbell

Timing and Structure

Fridays 9-11am plus afternoons, and Tuesdays 11-1pm

Prerequisites

3B2 (essential)

Aims

The aims of the course are to:

  • To become familiar with Verilog HDL and with the Lattice iCE40, a state-of-the-art low-power miniature FPGA used in many commercial embedded sensing and wearable computing applications.
  • To obtain experience working with FPGA synthesis tools for Internet of Things (i.e., embedded sensing and computing) applications.
  • To gain experience with the RISC-V architecture, an exciting, new, and forward-looking reduced instruction set computer (RISC) architecture.
  • To carry out the implementation and evaluation of a minimal subset of the RISC-V architecture on the iCE40 FPGA.

Content

This is a new project this year; some of the details below may evolve as the project content is developed further during the Michaelmas and Lent Terms.

Students will work in groups of three for this project.

The RISC-V architecture is a new open reduced instruction set computer (RISC) architecture that has many advantages over legacy architectures such as ARM. Because it was designed from the ground up for efficiency, RISC-V enables more efficient hardware implementations than many existing commercial architectures. One variant of the RISC-V is small enough to fit within the Lattice iCE40, a low-power miniature FPGA (in a 1.5x1.5mm package) targeted at embedded sensing / Internet of Things systems. This project will provide students with the opportunity to gain experience with the RISC-V architecture, an exciting, new, and forward-looking reduced instruction set computer (RISC) architecture and to implement and evaluate a minimal subset of the RISC-V architecture on the iCE40 FPGA.

Week 1

Complete the warm-up exercise mapping the provided single-instruction processor on the iCE40.

Week 2

Modifying provided RISC-V emulator to implement a new instruction.

Week 3

Mapping provided baseline RISC-V to the iCE40.

Week 4

Extending the provided RISC-V with the new instruction.

Coursework

Coursework Due date Marks

Interim report 1 (individual credit)

Report on analysis of the design and implementation of the provided single-instruction processor

4pm Friday 17th May 2019

(end of week one)

20

Interim report 2 (individual credit)

Implementation of modifying the provided RISC-V simulator to implement a new instruction and demonstration of provided baseline RISC-V design running on iCE40

4pm Friday 31st May 2019

(end of week three)

30

 

Final report (group credit for demo, individual credit for report)

Demonstration and report on modified RISC-V

4pm Friday 7th June 2019

(end of week four)

30

 

 

Examination Guidelines

Please refer to Form & conduct of the examinations.

Last modified: 25/09/2018 22:03

Engineering Tripos Part IIA Project, SF3: Machine Learning, 2018-19

Leader

Prof G Csanyi

Academic

Dr Richard Turner

Academic

Dr José Miguel Hernández-Lobato

Timing and Structure

Fridays 11-1pm and Tuesdays 9-11am plus afternoons

Prerequisites

Part I computing; Either of 3F3 or 3F8

Aims

The aims of the course are to:

  • expose students to machine learning approaches to non-linear regression and model-based reinforcement learning
  • to gain practical experience necessary to use these techniques successfully (e.g the use of training and test sets for evaluation, automatic differentiation for optimisation etc.)
  • to understand the robustness of these approaches to challenging real world phenomena including noise and non-linearities

Content

Note: This is a new project this year; some of the details below may evolve as the project content is developed further during the Michaelmas and Lent Terms.

 

In this project, students will consider the inverted pendulum system receiving a software simulator of a cart with a pendulum attached written in Python.
 
The goal will be to learn a controller that balances the pendulum in a data-driven way. The students will initially learn how to operate the simulator and explore the different types of behaviour that the system can exhibit. Next, they will collect training data from the simulator and use this to train non-linear regression models, including linear regression with non-linear basis functions. The trained models will be assessed on test data from the simulator. Once accurate models are learned these will be used to learn controllers that can balance the pendulum in the upright position and keep it there. Data-efficient model-based reinforcement learning techniques will be used for this stage. Finally, the controllers and the models will will be stress tested in various ways to test their robustness. 
 
Students work individually for this project. 
 

Week 1

Explore the cart-pendulum system using the simulator. Understand the state space and the governing differential equations.
 

Week 2

Gather training and test data from the simulator for building models of the system and validating them. Fit various models and assess their quality.
 

Week 3

Define a function that maps from the system's state to control actions (the "policy"), optimise the policy to keep pendulum upright. 
 

Week 4

Stress-test control and learning systems in various ways. 
 

Coursework

Coursework Due date Marks

Interim report

4pm Sunday 19 May 2019

30

Final report

4pm Thursday 6 June 2019

50

 

Examination Guidelines

Please refer to Form & conduct of the examinations.

Last modified: 28/09/2018 20:36

Pages

Subscribe to CUED undergraduate teaching RSS