Connected　and　automated　vehicles　(CAVs)　can　improve　traffic　flow　characteristics　by　implementing　advanced　control　algorithms　and　real-time　coordination　mechanisms.　In　the　near　future,　when　fully　autonomy　cannot　be　achieved,　it　is　expected　that　human-driven　vehicles　(HDVs)　can　be　indirectly　controlled　in　a　manner　similar　to　CAVs　through　an　in-vehicle　information　system,　which　has　the　potential　to　further　stabilize　traffic　flow　and　foster　cooperation　in　mixed　traffic.　This　article　contributes　to　this　area　of　research　by　designing　an　in-vehicle　longitudinal　trajectory　guidance　system　for　human　drivers　and　evaluating　its　impact　on　driving　performance　and　driver　behavior.　Given　a　pre-defined　longitudinal　vehicle　trajectory,　the　system　displays　the　real-time　guidance　information　through　a　virtual　block.　The　system　was　tested　through　driving　simulator　experiments　with　a　total　of　35　subjects.　Two　different　traffic　scenarios,　open　road　and　dedicated　road,　with　random　disturbances　were　considered.　The　experimental　results　indicate　that　the　proposed　trajectory　guidance　system　can　effectively　promote　efficiency　and　reduce　environmental　impact.　Side　effects　regarding　driver　behavior　are　limited　and　overall　compliance　with　the　system　is　satisfactory.　In　unexpected　pedestrian　crossing　settings,　drivers　can　recognize　road　safety　hazards　and　perform　braking　in　time　to　ensure　safety　with　trajectory　guidance.　It　takes　3.63　s　on　average　for　guided　drivers　to　regain　a　position　in　virtual　block　after　an　emergency.　The　findings　in　this　paper　can　support　the　realization　of　some　trajectory-based　traffic　control　methods　in　mixed　traffic　considering　random　road　safety　hazards.