Robot-arm path-planning simulator
A robotics simulator for comparing collision-free manipulator path-planning methods under joint, workspace, obstacle, and motion constraints.
Project definition
Problem statement
A robot end-effector can reach a target through many joint configurations, but obstacles, self-collision, singularities, joint limits, velocity, acceleration, and clearance determine whether a path is usable.
The engineering problem is to implement and compare path planners fairly, convert a geometric path into a feasible trajectory, and verify every motion in simulation.
Project objectives
- Model selected serial manipulators, joints, links, tools, obstacles, and task poses.
- Implement forward kinematics, inverse kinematics, collision checking, and workspace validation.
- Compare selected sampling-based and graph or optimisation-based planning methods.
- Smooth and time-parameterise paths under joint velocity and acceleration limits.
- Measure success, path length, clearance, computation time, smoothness, and repeatability.
System design
System modules
Robot and scene modeller
Loads a prepared robot, joint limits, collision geometry, tool frame, obstacles, and start and goal states.
Kinematics engine
Calculates forward and inverse kinematics, Jacobian indicators, reachability, and joint-limit checks.
Collision checker
Tests robot-obstacle and self-collision with configurable safety clearance.
Planner workbench
Runs selected RRT, RRT*, PRM, or trajectory methods using saved parameters and seeds.
Trajectory analyser
Smooths paths, applies time limits, rechecks collision, and compares motion and planning metrics.
Methodology
System workflow
- 01Load scenario
The student selects a prepared robot, obstacles, tool, start state, and target pose.
- 02Check feasibility
Reachability, inverse-kinematic solutions, collisions, and joint limits are validated.
- 03Run planners
Selected methods run repeatedly with recorded seeds, budgets, and parameters.
- 04Build trajectory
Successful paths are smoothed and time-parameterised under motion limits.
- 05Compare and replay
The 3D simulator rechecks and displays paths, clearances, joint traces, and metrics.
Demonstration scenario
A six-axis arm must move a tool between two fixtures while avoiding a central obstacle. RRT and PRM run with repeated seeds, their paths are smoothed and timed, and the simulator compares success rate, clearance, path length, joint motion, and computation time.
Engineering
Technical architecture
- Web application
- Nuxt and Vue for task setup, 3D playback, joint charts, planner controls, and comparison.
- Planning API
- FastAPI for robot and scene data, planning jobs, paths, trajectories, metrics, and exports.
- Simulation engine
- PyBullet for robot models, kinematics support, collision queries, and verified trajectory playback.
- Planning engine
- Python and NumPy implementations or adapters for selected planners, smoothing, and time parameterisation.
- Reproducibility
- Versioned robot and scene models, fixed planner budgets, recorded seeds, and regression scenarios.
Testing
Evaluation
Evaluation measures
- Planning success rate across prepared scenes and repeated seeds
- Collision-free and joint-limit-valid trajectory rate
- Path length, minimum clearance, smoothness, and execution time
- Planning time and explored states under fixed budgets
- Final pose position and orientation error
- Robustness to obstacle, start-state, and target changes
System boundaries
- The project operates in simulation and does not command a physical robot.
- Collision geometry, actuator dynamics, flexibility, backlash, sensing, and controller behaviour are simplified.
- A simulated collision-free path is not automatically safe on real equipment.
- Physical use requires calibration, safety systems, risk assessment, and qualified supervision.
Included
- 01Robot, obstacle, and task configuration interface
- 02Kinematics, collision, planning, and trajectory modules
- 033D path, joint, clearance, time, and comparison visualisations
- 04Prepared robot models, planning scenarios, and results
- 05Complete source code in a private GitHub repository
- 06Complete project documentation with synopsis, abstract, methodology, robot and planning diagrams, simulation results, screenshots, and conclusion
- 07Setup and usage guide
Project record
No buyer information is collected on this page.
- Permanent project ID
- GP-ME-1O87YCP
- Catalogued
- 21 Aug 2026
- Completed
- Pending
- Verified
- Pending
- Demonstration
- Added when ready
Handover
After purchase
- 01Payment is confirmed
The project is marked unavailable and cannot be purchased again.
- 02Repository access is granted
The buyer's submitted GitHub account receives access to the private repository.
- 03The purchase record is delivered
The certification sheet is prepared from the reviewed buyer details and sent privately by email.