Mandatory Performance Requirements
| M.P.1 | The system will explore the garage and reach a target parking spot within 2t time, where t is the time it takes to reach the target spot via user teleoperation |
| M.P.2 | The system will create a map of the environment with a mapping error of less than 20 cm as compared to the ground truth map |
| M.P.3 | The system’s map-building will be robust over sharp turns, such that the SLAM pipeline will not break (lose track of correspondences) in 80% of trials where a 90 degree turn is present |
| M.P.4 | The system will discretely localize itself within 5 degrees in rotation and 50 cm in translation as compared to the ground truth location |
| M.P.5 | The system will detect lanes of correct length and position in the image in 80% of the frames |
| M.P.6 | The system will create exploration waypoints that lie inside the lane in 80% of trials |
| M.P.7 | The system will park itself in the target parking spot within 10 degrees and 75 cm of the target pose |
| M.P.8 | The system will build a semi-dense map of its environment in near-real-time (with a delay of less than 1 second between receiving sensor data and updating the map) |
Mandatory Nonfunctional Requirements
| M.N.1 | System’s exploration algorithm will be scalable to a full-size autonomous car |
| M.N.2 | Prototype will be inexpensive i.e., less than $5000 to fabricate and assemble |
Desirable Performance Requirements
| D.P.1 | System will avoid dynamic obstacles, stopping 100% of the time a dynamic obstacle is introduced in the lane within 1m from the front of the robot |
| D.P.2 | System will detect empty parking spots using raw sensor data (no AprilTags) with an accuracy of 80% |
| D.P.3 | System will detect and classify road signs with a Mean Average Precision score of 80 |
Desirable Nonfunctional Requirements
| D.N.1 | The system’s architecture and software framework will be well-documented |
| D.N.2 | The system will notify the user of the status and location of the car once it is parked |
| D.N.3 | The system will choose between several open parking spots based on a heuristic |