Robot Null-Space Obstacle Avoidance in Simulation

ISEF Category: Robotics and Intelligent Machines

Subcategory: Robot Kinematics  ·  Difficulty: Advanced  ·  Setup: University Lab  ·  Time: Full Year

The Hook

A robot arm can reach the same hand position in more than one way. That extra freedom can help it slip around obstacles without losing the target. In your project, you test whether null-space motion really lowers collision risk while keeping tracking error small. That turns a math idea into a clear robotics experiment.

What Is It?

A robot arm with 7 joints has extra degrees of freedom, so it can bend in different ways while its end effector, the hand part, stays on the same path. That extra motion is called redundancy. Think of it like a person walking through a hallway with one hand on a wall. You can shift your shoulders and elbows without changing where your hand ends up.

Null-space motion uses that extra freedom to do a second job at the same time as the main task. The main task might be following a target path. The second job might be avoiding an obstacle, staying away from joint limits, or reducing awkward arm poses. In your project, a repulsive potential is a math rule that pushes the robot away from obstacles, like an invisible force field. You compare that strategy with a 6-DOF arm, which has no spare degrees of freedom and must give up more path quality when obstacles appear.

Why This Is a Good Topic

This topic works well because you can change one clear idea, the control strategy, and measure a real outcome, like tracking error or collision distance. You also get to study a core robotics concept that shows up in real robot arms used in factories, labs, and surgical tools. A student can learn kinematics, control, simulation, and quantitative comparison without building hardware. The project rewards careful modeling and analysis, not just coding speed.

Research Questions

• How does null-space obstacle avoidance change end effector tracking error compared with a non-redundant baseline?
• What is the effect of obstacle speed on tracking error in a redundant 7-DOF arm?
• Does stronger repulsive potential gain reduce collision risk without causing large path deviation?
• To what extent do different target path shapes change the benefit of redundancy in obstacle avoidance?
• Which joint-limit avoidance weight best balances smooth motion and task tracking accuracy?
• How does obstacle placement relative to the arm base affect the advantage of null-space control?

Basic Materials

• Computer with enough RAM to run robot simulation software.
• Python 3 with NumPy and SciPy.
• Robot simulation environment that can load URDF models.
• Franka URDF model file.
• Matplotlib for plotting trajectories and error curves.
• CSV or spreadsheet software for storing trial results.
• Notebook for recording control parameters and trial conditions.

Advanced Materials

• Workstation or lab computer with strong CPU support.
• ROS or ROS 2 simulation stack.
• Gazebo, MuJoCo, or PyBullet for dynamic testing.
• Franka URDF and joint-limit files.
• Python control and optimization libraries.
• Motion planning or inverse kinematics package.
• ImageJ or other plotting-free analysis software only if you export trajectory images for comparison.
• Version control system such as Git for tracking code changes.

Software & Tools

• Python: Runs your control code, computes error metrics, and compares trials.
• PyBullet: Simulates the robot arm and obstacle interactions in a physics environment.
• Matplotlib: Plots end effector paths, joint motion, and tracking error.
• NumPy: Handles vector and matrix math for kinematics and control.
• Jupyter Notebook: Helps you document experiments, results, and parameter sweeps in one place.

Experiment Steps

1. Define the single task your robot must repeat, such as following one path or reaching one target set, so you can compare every trial fairly.
2. Choose the control variable you will change first, such as repulsive gain, obstacle distance, or redundancy weight.
3. Build a baseline model for the 6-DOF arm and a comparison model for the 7-DOF arm so you can separate redundancy effects from general control quality.
4. Decide how you will measure success, including tracking error, minimum obstacle distance, smoothness, and any joint-limit penalties.
5. Plan a small parameter sweep that tests several controller settings instead of one lucky value.
6. Set up a fair analysis method that averages repeated runs and compares the two arms with the same target path and obstacle motion.

Common Pitfalls

• Comparing the 7-DOF arm to a different target path than the 6-DOF baseline, which makes the result unfair.
• Tuning the repulsive potential so aggressively that the arm avoids obstacles but stops tracking the task well.
• Ignoring joint limits, which can make the null-space strategy look better than it really is.
• Measuring only collision avoidance and skipping tracking error, which hides the cost of the control choice.
• Changing simulation settings between trials, which can shift results because of solver or timestep differences.

What Makes This Competitive

A stronger project goes beyond one demo run. You would test several obstacle layouts, several target paths, and several controller gains, then compare the patterns with real statistics. You can also ask whether the redundant arm wins only near obstacles, or whether it helps even when the path is easy. Careful control of baselines, clean plots, and a clear explanation of tradeoffs can make the project feel research-level.

Project Variations

• Test the same null-space controller with a different 7-DOF arm model, such as another industrial manipulator, to see whether the result depends on arm geometry.
• Replace the moving obstacle with a static obstacle field and compare how much redundancy helps in steady versus dynamic environments.
• Add a second metric, such as joint smoothness or energy-like motion cost, to see whether obstacle avoidance creates less comfortable arm motion.

Learn More

• MIT OpenCourseWare: Search for robotics, kinematics, and control lecture notes that explain Jacobians, inverse kinematics, and redundancy.
• NASA NTRS: Search for robot arm and space robotics papers that discuss redundancy and obstacle avoidance in manipulator control.
• PubMed: Search for review articles on robot motion planning in surgical and rehabilitation robotics if you want human-facing examples.
• IEEE Xplore: Search for journal papers on null-space control, redundancy resolution, and potential field methods through your school or library access.
• ROS Wiki: Read the open documentation on robot models, controllers, and simulation tools used in research and testing.