Development and Analysis of a single leg for a Quadruped Robot

Project Title

Development and Analysis of a single leg for a Quadruped Robot

Project Area of Specialization RoboticsProject Summary

Problem Statement: 

Land animals running at high speeds need high accelerations and their legs experience very high loads. These high loads are faced by their structure and are also endured by their muscles. The robotic counterparts of these animals face these issues and demand the design of robotic legs to have good strength but with low weight.

Proposed Solution:

To address the afore mentioned problems nature provides us examples like the Cheetah or the Greyhound, which portray excellent characteristics in terms of leg design ,actuation and its control. These animals utilize their muscles which absorb high impact forces and prevent the bones from breaking. We have adopted a similar mechanism in our design which enables a lightweight, robust and endurable structure.

The Reason:

Our FYP is to develop and analyze a single robotic leg which will serve as a basis to the knowledge on which the viability of the mechanical quadruped (four legged robot) will be questioned. Once the testing is completed, we can replicate our project and make the remaining legs for the quadruped. This would save us cost and would prove to be a more efficient process.

Ultimate Goal:

The ultimate goal is to make a functional robotic quadruped which will be the second phase after the completion of our project. This quadruped is intended to be able to perform various motion gaits (e.g. “walking” or “trotting”) on uneven surfaces with great efficiency and would also be able to carry payloads.

CAD design:

 

Project Objectives

Project Objective:

The primary objective is to make a fully functional single leg for a quadruped and to analyze its design. This will aid in making a complete quadruped in the future. Characteristics such as endurance, strength, rigidity, impact mitigation and a lightweight structure found in animals are aimed to be incorporated in our project.

The secondary objective is to bring the efficiency of the leg as close as possible to an actual animal. This is governed by the amount of energy stored and recuperated in general robots. However, we aim to minimize the energy application through control algorithms coupled with the conventional goal gets us closer to the efficiency to an actual animal.

Lastly, present quadruped robots are very expensive and an average person cannot afford to work on it. Hence, we aim to build a cost effective structure on which students and researchers can test their findings or even recreate at their own disposal.

Project Implementation Method

Some key Parameters: 

• We will be developing a mechanical structure for the leg (ensuring minimum distal leg weight without compromising strength, by performing topology optimization and FEA).
• Control Algorithyms that enable the leg to move in desired trajectories and detect surfaces and height drops through multiple sensors (paths, trajectories and controller modelling using Matlab to interface with Labview and MyRio.)
• An original Series Elastic Actuation (SEA) that ensures proper impact mitigation and energy recuperation. This protects crucial components from impact forces and makes the system more efficient. 
• A modular and lightweight structure coupled with high torque NEMA 23 Brushless DC motors equipped with hall sensors ensure accurate to 0.5 degrees. 
• Exclusive sensors measure the torsional spring deflection (a part of SEA) at the time of impact and the force faced by the foot. 

Benefits of the Project

Benefit of our project: 

Our project is made in scope of the goal of achieving a functioning quadruped in the future. The benefits we achieve by carrying out project are:

• The viability of the structural design is tested.
• Adequate impact mitigation is found through experiments on the leg.
• Cost effective, since there is only a single leg in question.
• Focus is primarily on the functioning of a single leg, which enables prototyping to be more diverse.
• Creating a control model for one leg is easier, than the direct approach of an entire quadruped.

Ultimate Benefits:

This type of robot has many applications such as:

• - The armed forces may use it to carry packages to places where it’s too dangerous or where wheeled/tracked vehicles cannot go.
• - Nuclear researchers often transport radioactive materials that require delicate handling can use our robot.
• - Minefield surveyors come across uneven terrain where the excavation process requires power and accuracy.
• - Our robot can be a base structure on which AI researchers can validate and further carry out their experiments
• - First response teams can send in the robot to places such as a compromised building to aid people or retrieve valuables.
• - Demolition organizations often require checks to be made that are too risky for actual humans
• - Area Mapping Personnel can get access to a more vast area through our robot.

Technical Details of Final Deliverable

We have divided the Deliverables in two categories which indicate the intensity of requirement (D = Demand, W = Want)

</

1	Functional (Attributes that relate to project’s functionality)
	D/W	Description
1.1	D	Demonstrate a simple gait carried out by actual quadrupeds i.e. Walk.
1.2	D	Capable of performing gaits under 20 N vertical load.
1.3	D	Impact Mitigation of ground forces transferred through limbs to the body of the robot.
1.4	D	To jump a vertical height of at least 100mm.
1.5	W	Perform motion with a vertical load of 30 N.
2	Physical (Attributes that relate to project’s nature)
2.1	D	Structure capable enough to withstand 20 N vertical load.
2.2	D	Low weight of the structure to not exceed more than 8kgs.
2.3	W	Low inertia design. Minimum weight at distal leg.
3	Design and Construction (Attributes that relate to project’s overall design and manufacturing)
3.1	D	Fabrication of testing rig which enables and accesses all features of the single leg and the construction of the leg itself.
3.2	D	Development of a control system for the quadrupedal leg.
3.3	D	Mathematical model of the “joint torque estimation”
3.4	D	3D Modelling of the quadrupedal leg.
3.5	W	A Modular design that enables easy maintenance.
4	Environment (Attributes that relate to project’s operating conditions/ impact on environment)
4.1	D	Noise Pollution should be less than 50db.
4.2	D	Electrically actuated to avoid any toxic emissions.
4.3	W	User Friendly.
		<

1

1.1

1.2

1.3

1.4

1.5

2

2.1

2.2

2.3

3

3.1

3.2

3.3

3.4

3.5

4

4.1

4.2

4.3

Final Deliverable of the Project HW/SW integrated systemType of Industry Education , Manufacturing , Transportation , Media , Health , Security Technologies Artificial Intelligence(AI), Robotics, OthersSustainable Development Goals Good Health and Well-Being for People, Industry, Innovation and Infrastructure, Life on LandRequired Resources

Elapsed time in (days or weeks or month or quarter) since start of the project	Milestone	Deliverable
Month 1	Literature Review	Acquired multiple designs that we could work on.
Month 2	1) Mathematical model for Joint Torque Estimation. 2) Initial Sketch of CAD model.	1) BLDC motors selected and ordered.
Month 3	1) CAD model finalized. 2) SEA design finalized. 3) Path and trajectory modeled.	1) All required parts known. Mechanical limitations known. 2) Design of spring decided and evaluated. Deflections and Load ratings known. 3) Graphs for the working area for the end effector to work in determined.
Month 4	1) FEA and Topology Optimization carried out. 2) Detailed design report finished. 3) Desired Sensors selected	1) Estimated weight finalized. Compared with BLDC motors to get torque to weight ratio. 2) Hardcopy of all our calculations reviewed and approved. 3) Sensors ordered.
Month 5	1) Leg Structure material acquired 2) BLDC motors arrived.	1) Selected Parts sent off to 3D printing or CNC machining. 2) Initial tests carried out on the motors. Everything checks out.
Month 6	1) Custom Power transmission manufactured. 2) Upper and lower joint pins manufactured. 3) Custom double torsional spring made.	1) Gears tested with motors.
Month 7	1) Sensors arrived. 2) 3D prints completed. 3) Algorithms finalized	1) testing performed on sensors. 2) Finishing and testing performed on the 3D parts. 3) simulation graphs show desirable and stable results.
Month 8	1) CNC parts finished 2) transfer function of BLDC motors calculated for integration with software. 3) All parts manufactured, welding and minor fittings remain.	1) Lower leg completed 2) Motors controlled through software. 3) Labview, Myrio and BLDC interface created.